You know what, Google forsakes many sites fairly frequently, even sites that a lot of people like. Why would MetaFilter be spared? Because it's popular with techies and Google can do no evil in their eyes?
Disruptions, unpredictability, and random swings in support are all harmful to the ecosystem. Google's actions have consequences.
Alternatives are to make yourself far less dependent on Google (all the more reason to support search alternatives, including DDG). But so long as your business is eyeball aggregation, you're going to be at the whim of the aggregators (and that is a control point).
I don't disagree with what you said, but it's constantly pointed out that start-ups (etc) should avoid being dependent on x y z for traffic.
There isn't a single business on earth that is not dependent on a 'platform' of one sort or another. Even Google is dependent on other platforms. You merely pick your poison, if you're so lucky, when you run a business. There are always gatekeepers and always will be; and they can cripple you if they choose to, the only variable is how much of an incentive they have to do so, and the damage they'll take by doing so.
Make yourself less dependent on Google? Ok, which means you're going to make yourself more dependent on Facebook, or Twitter, or LinkedIn, or DuckDuckGo, or radio ads, or television ads, or newspaper ads, or billboards, or etc. None of those are benevolent lords just looking out for your own well being. DDG can still chop you up with an algorithm change just as easily as Google can, and at Google's scale every little change would impact businesses just as Google's present changes do, and there would be just as many claims of unfair treatment. DDG can be as transparent and open as they like, that wouldn't change the sour grapes that would be derived from algorithm changes.
You're omitting another option: invest in a business opportunity which isn't dependent on online visiblity at all. If you're in the mass-consumer space, eyeballs matter. If you're in a more selective space, the number of eyeballs matters _far_ less than their quality. A B2B proposition cares about exposure within their chosen industry, a neighborhood or regionally-oriented business about local word-of-mouth or reputation. A consultant really only needs sufficient clients to keep booking solid. Etc.
Google might play a role, but odds are far higher that other methods: cold calling, networking, flyers, street signage, radio or TV advertising, industry-specific publications publicity, etc., play a far larger role.
The problem with the online space is that, if you're going to play there, Google owns something on the order of 80-85% of all search. There are other avenues to eyeballs (Facebook, Twitter), but there are specific components to their reach.
"Ok, which means you're going to make yourself more dependent on Facebook, or Twitter, or LinkedIn, or DuckDuckGo, or radio ads, or television ads, or newspaper ads, or billboards, or etc."
Being dependent on Google + Facebook + Twitter + LinkedIn is better than being dependent on Google alone.
MetaFilter is just now learning that it did not have a sustainable business model.
As a MeFite of several years, I've gotten a lot of useless/terrible advice through Ask MeFi. Basically for any subject remotely esoteric, you'll get a lot of "subject 101" answers and ones that are totally wrong. For general/life advice it's pretty good though, mostly because of the number of older users there.
Everyone on there is a 30-something linguist, musician, or grad student; is an experienced fetishist; is a traveller who goes off the beaten track; is involved in several local community organizations; and has a long history of therapy.
Metafilter's founder mathowie takes a much more measured tone about Google's role: "A year and a half ago, we woke up one day to see a 40% decrease in revenue and traffic to Ask MetaFilter, likely the result of ongoing Google index updates." (http://metatalk.metafilter.com/23245/State-of-MetaFilter)
Google changes. The web itself changes. As seminal as Metafilter has been, it doesn't "deserve" the top slots any more than any other content.
So I find it a little irritating that David Auerbach at Slate has attempted to overdramatize this into something it's really not.
Even if the prior position were unearned, the point that, perhaps, small changes in Google's algorithms can have dramatic influences over site rankings, with all the concomitant disruption that can entail, argues for conservatism on Google's part. Otherwise you're in the position of the philanthropist who, out of the goodness of his heart, supports some cause unreservedly ... until he doesn't. Suddenly withdrawn support has very real impacts.
If there _was_ a specific strike against Metafilter, Google really ought explain. Actually, even if there wasn't.
And it argues again against digital sharecropping, though that's harder and harder to avoid.
I think Auerbach laid out some cogent reasons why this should be questioned. A gradual decrease over time? Sure, expected. A sudden drop? That's an algorithmic hit, and when a site with good content (that is ranked well by other search engines) is hit negatively by the algorithm, it should be questioned.
And, unfortunately, the only way to get Google to consider such a question is if enough people start echoing it.
What do you mean it doesn't "deserve the top slot any more than any other content"? That doesn't make any sense at all. Obviously if it is the "best" answer to the search, then of course it "deserves" a high result. The author lists out some examples where this doesn't seem to be the case.
I didn't find the article overdramatizing at all. In fact it was quite measured and completely reasonable. And the timing is obviously consistent with Metafilter's recent announcement.
That a high quality site gets chopped in half doesn't bother you at all? And that Google's results seem off? Really, this doesn't deserve a mention?
What's the justification that metafilter is in fact the best result for some wide range of queries? The content these days doesn't look drastically superior to Reddit / Stack Overflow etc. The author mainly argues the value of the content by citing threads from 2004 to 2009.
In order to make the point that Google has done something wrong here you need to demonstrate that there's actively superior content over there compared to the more mainstream communities, and the author doesn't persuade me.
The content these days doesn't look drastically superior to Reddit / Stack Overflow etc.
There is zero spam, zero meme threads, an engaged and involved membership and, perhaps a glaringly-missing signal on Google's part, consistency over a long time.
But even more than that, why would Ask Metafilter threads about e.g. renovating a kitchen rank lower than the 12 effectively-identical content-farm "howtos" from About, WikiHow, et al, that don't involve real people describing real experiences and knowledge? It's really a matter of prioritization, of what Google considers "superior," and I think there are some positive qualities that Google doesn't account for. Metafilter is certainly no worse for the Internet in general than Stack Exchange, doityourself.com, rottentomatoes, or other commonly high-ranked sites.
>There is zero spam, zero meme threads, an engaged and involved membership and, perhaps a glaringly-missing signal on Google's part, consistency over a long time.
I don't think this is drastically different from any of their competitors except the Reddit defaults, which are pretty much irrelevant for these purposes anyways (excluding consistency, which I don't really think is relevant when you're searching for something on Google).
>But even more than that, why would Ask Metafilter threads about e.g. renovating a kitchen rank lower than the 12 effectively-identical content-farm "howtos" from About, WikiHow, et al, that don't involve real people describing real experiences and knowledge?
I think About and WikiHow could lack the depth that Metafilter has, but they're going to win out on accessibility. I think a dedicated subreddit would win out on both fronts. If Google finds that people prefer accessibility, then that's what Google will deliver. I don't think there's anything wrong with that.
The issue is that it's ranking below extremely low quality sites in many cases, not that it's ranking below SO or Reddit (and in my experience, Metafilter results are almost always higher quality than Reddit threads, less so for SO, but those are only technical whereas Metafilter represents many more topics)
...it doesn't "deserve" the top slots any more than any other content.
I thought the whole deal with inventing Page Rank and all these smart indexing algorithms - the whole raison d'être for Google - was exactly because some content was more deserving than other content.
When I do the search for "maybe there is a god," MeFi doesn't show up until page 8 now. The first two pages of results are virtually all videos and lyrics related to a single 12-year-old Christian pop song.
One thing I wanted to mention earlier, didn't have time:
The pop song that comprises the first two pages of results for the "maybe there is a god" search is actually called "Mabye There's A Loving God." That's close to the search text, but it's not an exact match.
The MeFi page mentioned in the article contains the exact phrase "maybe there is a god" in the title. And it shows up on page eight of my search results. That is ridiculous.
You can argue that MeFi wasn't singled out on purpose, that algorithms change and there are bigger benefits elsewhere that I'm not seeing in this case and blah blah blah, and I don't know about any of that. No opinion.
But I think you're going to have a hard time arguing that these search results are in any way more accurate, more relevant, or better than the older results were.
And that is very much of a piece with recent, more general criticism of Google's search: That it's simply not as good, not as useful, as it used to be.
But this isn't just about MetaFilter dropping out of the top slots. Google is actively accusing other sites of spamming because MetaFilter users linked to them. Google has clearly miscategorized MetaFilter as a spam farm, and is penalizing it and others because of this algorithmic error.
I think Metafilter needs to spend more time exploring SEO and perhaps a site redesign. It may be a great site (people say it is), but if it has a poor backlink profile, stale content, and a high bounce rate, it's not going to rank as well even for better content.
Craigslist is a great site too, and you don't see it highly ranked in Google for many queries.
absolutely, the website looks very 90s, none of the links look like links and its too myspace in appearance (old myspace) - I would bounce if results landed me on that site without even reading the content.
I try to avoid it because it looks like garbage. Even though the content is good, I feel the site is not authoritative because it looks like some guy's geocities page.
I agree that the design is a bit old-fashioned, but to me sites are about the content, so as long as I can get the information I want without getting a migraine, I'll ignore it. Authoritative comes from content, not looks.
In fact, to me MeFi and HN are about the same, design-wise...
A high bounce rate is unlikely to produce the massive dropoff shelf in their traffic profile. They have likely tripped some flags and are getting a site penalty.
The fact that Metafilter hasn't jumped on a bunch of javascript-heavy design fads over the years is one of the things I really like about it. Much like Hacker News, the site just works and doesn't get in your face with a lot of designer ego.
No reason to jump on "javascript-heavy" design fads, a simple refresh of the design to be more modern and usable would suffice. Even reddit is more appealing and visually straightforward for new users, and that site usually ranks pretty low for readability for new users unaccustomed to the design.
And this is why I got out of organic SEO. You have essentially one client, and that is El Goog. A site can see product sales revenue drop 50% overnight after a bad Google dance, and it can take weeks to work out how you've offended it. I only used whitehat techniques mind.
Most people bleat "just write good conteeeeent" and have no idea how Google ranking actually works. I guess what was most galling was seeing certain retail sites getting away with blatant spam while your own sites got mysteriously slammed.
I don't blame Google though - it's a direct result of having a single player market. So go, duckduck, go :)
Especially now that Google is a "content provider," sites like Metafilter have to battle Youtube, Google+ etc for ranking. Guess who's winning? Traffic is going away from other sites to Google's own properties and there's isn't a damn thing metafilter et al can do: Google handles ranking, display and ads.
Commercial sites now have to deal with a full page of ads before the real results so their fate is sealed another way. In other words, we need 4-5 search engines, not one giant monopoly that seeks to increase ad clicks by underhanded methods each quarter.
I sifted through a lot of traffic data, and it must be noted that I saw zero evidence of El Goog manually weighting any results. Everything was 100% above board - they might donkey-punch you, but it was algorithmic and not manual.
I bet MeFi could be fixed in a few months if they hired a real SEO expert, ie someone who understands the underlying ranking algos and not just the usual voodoo bullcrap. The site is probably just tripping a few red flags, it could be something as stupid as an accidental link loop in their dynamic content.
I'll give it to Google, once you find your sin and fix it, your traffic will come back, maybe not 100% but a significant recovery. (Unless you have done something really dumb/blackhat like paid links.)
a link loop is a red flag?
I guess I am misunderstanding, as that would mean I can't have a "artticles like this <link> <link>" block in my blog, which seems very weird.
A link loop as in a dynamically generated menu structure that creates an endless loop for the crawler, or accidentally makes large numbers of very similar pages with similar URLs, etc. (Yes this can happen legitimately and by accident on retail/commercial sites).
I haven't used one of those myself, but yeah that does sound dangerous - however you could just tell the Googlebot to ignore the link in GA and it should be fine. I've seen people do similar things to create infinitely deep directories to screw with evil crawlers, since the Googlebot should obey your directives.
> they might donkey-punch you, but it was algorithmic and not manual.
This is not always the case. Look at the Rapgenius event. That was triggered by manual intervention.
>I bet MeFi could be fixed in a few months if they hired a real SEO expert, ie someone who understands the underlying ranking algos
There isn't anyone on the outside that truly understands them because they constantly adjust them to deal with abuse. It is possible that one of these adjustments killed MeFi and it wasn't a recent mistake on MeFi's part.
You are correct that manual penalties can be applied, I worded my previous comment badly. What I mean is they don't appear to "manually" tamper with the result weightings eg by using a whitelist of upweighted domains.
While they can slap a manual -50 penalty on your domain, from memory you are notified of such penalties in Google Analytics so it's not a silent and nefarious manipulation. MeFi should know if they have such a penalty. I also admit that the "get out of jail" process is even more annoying than dealing with an Apple app store rejection.
You miss the point completely: If Google ads an extra ad you might lose 20% of your traffic even if your rank stays the same. Or if they ad youtube videos on the display, or books, or local or whatever....
So call it manual, algorithmic, display change or whatever, you're f-ed and call moves more of web traffic to its own properties.
>>One last one: Searching for “most amazing woman ever” on Bing will give you MetaFilter’s helpful “Who is the most amazing woman who ever lived” as the third result.
>>Google puts it at the bottom of the second page of results, in 19th place. Google’s top result? A list of “the 100 most beautiful women ever.”
Unfortunately, American culture (and many cultures around the world) objectify women and value them based on their looks. So Google's search results may just be a true reflection of what people are interested in looking at. Perhaps Bing's algorithms aren't as good as Google's algorithms in estimating what people might be more interested in.
To be clear, I didn't intend my statement as a feminist critique of society (though personally I would agree with this feminist critique). I just used this example as a roundabout way of defending Google's algorithm and saying that Google gives us the search results we deserve (and ask for) and not the search results that (from Auerback's perspective and mine) we should probably be more interested in.
That's the crux of the argument, though: Google search results are ranked according to a value system. What does that system prioritize? Not Metafilter; not Harriet Tubman. "Beautiful Women." This is neither right nor wrong, it just is, but Google has to live with what that says about them, and that's entirely fair.
"Searching for “most amazing woman ever” on Bing will give you MetaFilter’s helpful “Who is the most amazing woman who ever lived” as the third result. (Answers, by the way, included British spy and French Resistance leader Nancy Wake, world’s first programmer Ada Lovelace, slave rescuer and activist Harriet Tubman, and Chinese pirate Ching Shih.) Google puts it at the bottom of the second page of results, in 19th place. Google’s top result? A list of “the 100 most beautiful women ever.” "
That made me wonder whether Google has shifted its products from being the kind of thing that their own employee corps would enjoy to things for the general public. I personally f.i. enjoyed the from: to: keywords in Google Maps search. That feature disappeared and probably the majority of users wouldn't use those anyway.
Personally I scorn at a search result of the '100 most beautiful women'. But in the large population it will probably lead to a much higher click through.
So I wonder if Google Search was oriented towards the smart crowd niche and that they changed focus towards the common and the populous.
To put it differently: a shift from an emphasis in information as high value search result towards including entertainment, gossip etc. High brow to low brow.
This is by far the best reason I've heard for the decline in quality of Google search results. There's no doubt that when I first started using Google it was for technical things, and I feel like most of the people online during their first few years of life were doing the same. I'm guessing we're no longer their target market, although I wouldn't be surprised if an algorithm made that decision instead of a person.
At the same time, they are trying to do targeted search and I feel like that should be able to help far more than it is.
> I personally f.i. enjoyed the from: to: keywords in Google Maps search. That feature disappeared and probably the majority of users wouldn't use those anyway.
I still use that feature every day. I have a smart keyword setup so that I just type:
maps from LAX to San Francisco
into the URL bar and get taken straight to a Google Maps page showing me that route [1]. Forgive me if I've misunderstood you and you're talking about something else, but this feature still exists. AFAIK I'm using the latest version of Google maps.
In classic Google Maps you could find a multi-leg route by searching with the syntax "from:san francisco to:los angeles to:new york" [1] whereas in new Google Maps the same query just shows you San Francisco [2].
The new Google Maps will do multi-point routes [3], but by default when you enter San Francisco to Los Angeles it chooses flights, and hides the option to enter a multi-point route. You have to switch to car directions to get the option to enter more than one point, and enter them one at a time not use the nice from-to-to syntax.
Pretty crap for power users - but I guess ad-supported products make their money by being mass market, and catering to power users isn't very important.
It also bugs me that there's no way to enter the mode of transportation as part of the URL query string. I'd love to be able to use smart keyword syntax like:
from London to Cambridge by bike
from Glasgow to Manchester by bus
Another example would be how google did sunset Google Reader. That also struck me as a product that was aimed at people who are rather similar to Google employees.
where is the growth, man? and i don't mean the dropoff. the dropoff is just, you know, change. google changes, all the time, user behaviour changes, user expectation changes, the way google treats sites changes, the internet changes. so yes, 40% less traffic via google organic search happens. you don't have any "right" to google traffic, there are a lot of sites out there.
but the thing that seems to be really broken is the "lack of growth" before the dropoff and after it.
SEO overall has these business cases
* a) if we add x% more landingpages to our site, we gain y% more traffic.
* b) if we add x% more content to y% of our landingpages, we gain z% more traffic.
* c) if other sites point x% more links to our landingpages, we gain y% more traffic.
* d) if we make the user experience of our site better, our traffic gets more sustainable.
as metafilter is a quite popular site with an active community, they should at least do a) and b) quite well, probably doesn't have an issue with c)
as there is and was not growth, then there is probably something rotten going on on the platform level.
so basically before firing their staff they should have listen to a decent SEO (and implemented some changes) and / or mobile guy.
tl;dr: if your have a popular site with lots and lots of content, but you don't see (organic search) growth then your site is broken, even before google turns of some of your traffic because you know, google.
p.s.: i would love a look into meta-filters google webmaster tools
I'll be honest - I've been avoiding MetaFilter for years due to the horrific UI/design.
I actually didn't know that the content MetaFilter has supposedly is good because the design is so off-putting that I've never been able to read much of it.
MetaFilter should really do a redesign to a more readable and friendly interface, that is likely to increase usage/reduce bounce rates (and maybe increase the Google rankings).
I'm far from a designer, but I still like to think I appreciate good design.
1. The color combinations are really bad. Dirty-yellow on blue/dirty-yellow on green is close to what a designer would chose if they were planning to make something deliberately look ugly.
2. The titles are in grey while the content text is in white. Why tone down the most important thing on the page?
3. The spacing between all the objects is weird. The space between the comments is huge - it was likely made that large since they would otherwise flow into each other. They should have separated the comments using other methods in addition to just making the space between them huge.
4. For long comments with multiple paragraphs and links it can be hard to see where the next comment begin, since the links (in the comment) and the links to the user profile/permalink to post look almost the same.
I am not a designer, but I know that I do appreciate good design.
Even back in the early 2000's, Metafilter's simple, clean, relatively information-dense design was a welcome change from most news aggregators out there. Metafilter is supposed to be a place for thoughtful medium-to-long-form discussion. I feel that its clean, consistent, text-focused design serves that purpose well.
Honestly, remove comment nesting, move the comment credit line under the comment, and change the color scheme to white and gold on blue, and HN's comment page looks pretty much exactly the same as Metafilter's.
Your last paragraph basically says "remove the biggest flaws with Metafilter's design compared to HN's - and then they are pretty much the same". ;)
I won't go into the color scheme since I guess that is a matter of taste, but at the same time I don't think it's an accident that you don't see the same color combinations used anywhere else ;)
The comment nesting and placement of the comment credit line(s) is actually really important. It is vital to make the different comments easy to visually separate from each other. HN has one line above and one line below each comment, effectively boxing them in. That, plus the comment nesting, makes HN's comment threads easy to read and skim over.
Metafilter on the other hand has just one line below each comment. Unlike on HN, where the line has a lower contrast/eye focus than the comment text, the line on MF is using the same colors as the links in comments. That is completely counter-productive, and objectively bad design.
Since MF doesn't have comment nesting it should do something else to visually separate the comments. Most other sites subtly change the background color, but there are also other ways to do it.
High contrast/big size should be used for things that should be focused on, and small contrast/small size for things that shouldn't be focused on. On HN the contrast between the color and size of the objects mostly make sense (although self posts should have darker text), but on MF it is all over the place.
Like HN, I feel that MeFi keeps things simple because it doesn't want its community to grow linearly. Which is great, because the bigger a community gets, the less friendly and manageable it becomes.
I can't help feeling that there's something wrong when the only way to find out why Google has downgraded your site is to attract the attention of one of the core search developers. I get that Google wants to keep it's ranking algorithms secret for fear of black hat SEO poisoning the well even more than it does already. Given Google's dominant position in the search market at the same time it seems wrong to downgrade sites in this fashion with no clarity or visibility over the process whatsoever.
Wow: That really was a pile on wasn't it? Looks like mcutts is catching some of the backdraft from Google's infamous lack of customer service (to paying advertisers, content providers and everyone else) here to a certain extent.
Well Matt is a relentless explainer of Google's position. That earns him a lot of credibility and support when the position makes sense, but if what they are doing ultimately is considered harmful, then he's going to look like an apologist.
mcutts is pretty much the only thing resembling customer service for google search, so he's going to catch all of it. while mcutts has put himself into this position willingly, it's google's fault for making him a solitary target.
Hmmm... so Google is becoming less relevant to me. When I search for content, I expect to see more MetaFilter than "100 most beautiful women in the world".
Duck Duck Go... perhaps it's time to switch my search engine?
I feel like a few people are missing the point as to why MetaFilter dropped in their rankings. It was likely not a manual penalty but rather a change in regards to the Panda algorithm. According to this article (https://medium.com/technology-musings/941d15ec96f0) the drop happened in November of 2012. It doesn't say which date, but November 5 and November 21 were both days that Google refreshed the Panda algorithm.
Panda was created to filter down poor content and filter up good content. Sometimes it gets it wrong, but generally if a site is being filtered down it's because there are too many pages of the site in the index that are apparently not helpful to users.
I do think that the design has a lot to do with MetaFilter's demise. I have landed on this site, and, not knowing what I know now after reading articles about the quality information that is on MetaFilter, I have immediately clicked away because it looked like a spammy page. If enough users do this then this shows Google that users don't want to engage with this site.
I haven't had a deep look into the technical structure of the site but there's a good possibility that with a redesign and a review of their architecture, the site could see a return to normal levels with the next Panda refresh.
Full disclosure: MetaFilter member reading about this stuff everywhere with interest who joined here to comment. HN seems right up my alley with the simple design though I'm not sure if I'm ever going to be a threaded comments kind of guy.
81 comments
[ 3.2 ms ] story [ 120 ms ] threadI'd be interested in seeing a more rigorous data set.
Keyword Phrase, bing ranking, google ranking
Alternatives are to make yourself far less dependent on Google (all the more reason to support search alternatives, including DDG). But so long as your business is eyeball aggregation, you're going to be at the whim of the aggregators (and that is a control point).
There isn't a single business on earth that is not dependent on a 'platform' of one sort or another. Even Google is dependent on other platforms. You merely pick your poison, if you're so lucky, when you run a business. There are always gatekeepers and always will be; and they can cripple you if they choose to, the only variable is how much of an incentive they have to do so, and the damage they'll take by doing so.
Make yourself less dependent on Google? Ok, which means you're going to make yourself more dependent on Facebook, or Twitter, or LinkedIn, or DuckDuckGo, or radio ads, or television ads, or newspaper ads, or billboards, or etc. None of those are benevolent lords just looking out for your own well being. DDG can still chop you up with an algorithm change just as easily as Google can, and at Google's scale every little change would impact businesses just as Google's present changes do, and there would be just as many claims of unfair treatment. DDG can be as transparent and open as they like, that wouldn't change the sour grapes that would be derived from algorithm changes.
You're omitting another option: invest in a business opportunity which isn't dependent on online visiblity at all. If you're in the mass-consumer space, eyeballs matter. If you're in a more selective space, the number of eyeballs matters _far_ less than their quality. A B2B proposition cares about exposure within their chosen industry, a neighborhood or regionally-oriented business about local word-of-mouth or reputation. A consultant really only needs sufficient clients to keep booking solid. Etc.
Google might play a role, but odds are far higher that other methods: cold calling, networking, flyers, street signage, radio or TV advertising, industry-specific publications publicity, etc., play a far larger role.
The problem with the online space is that, if you're going to play there, Google owns something on the order of 80-85% of all search. There are other avenues to eyeballs (Facebook, Twitter), but there are specific components to their reach.
Being dependent on Google + Facebook + Twitter + LinkedIn is better than being dependent on Google alone.
MetaFilter is just now learning that it did not have a sustainable business model.
Google changes. The web itself changes. As seminal as Metafilter has been, it doesn't "deserve" the top slots any more than any other content.
So I find it a little irritating that David Auerbach at Slate has attempted to overdramatize this into something it's really not.
If there _was_ a specific strike against Metafilter, Google really ought explain. Actually, even if there wasn't.
And it argues again against digital sharecropping, though that's harder and harder to avoid.
And, unfortunately, the only way to get Google to consider such a question is if enough people start echoing it.
I didn't find the article overdramatizing at all. In fact it was quite measured and completely reasonable. And the timing is obviously consistent with Metafilter's recent announcement.
That a high quality site gets chopped in half doesn't bother you at all? And that Google's results seem off? Really, this doesn't deserve a mention?
In order to make the point that Google has done something wrong here you need to demonstrate that there's actively superior content over there compared to the more mainstream communities, and the author doesn't persuade me.
There is zero spam, zero meme threads, an engaged and involved membership and, perhaps a glaringly-missing signal on Google's part, consistency over a long time.
But even more than that, why would Ask Metafilter threads about e.g. renovating a kitchen rank lower than the 12 effectively-identical content-farm "howtos" from About, WikiHow, et al, that don't involve real people describing real experiences and knowledge? It's really a matter of prioritization, of what Google considers "superior," and I think there are some positive qualities that Google doesn't account for. Metafilter is certainly no worse for the Internet in general than Stack Exchange, doityourself.com, rottentomatoes, or other commonly high-ranked sites.
I don't think this is drastically different from any of their competitors except the Reddit defaults, which are pretty much irrelevant for these purposes anyways (excluding consistency, which I don't really think is relevant when you're searching for something on Google).
>But even more than that, why would Ask Metafilter threads about e.g. renovating a kitchen rank lower than the 12 effectively-identical content-farm "howtos" from About, WikiHow, et al, that don't involve real people describing real experiences and knowledge?
I think About and WikiHow could lack the depth that Metafilter has, but they're going to win out on accessibility. I think a dedicated subreddit would win out on both fronts. If Google finds that people prefer accessibility, then that's what Google will deliver. I don't think there's anything wrong with that.
I thought the whole deal with inventing Page Rank and all these smart indexing algorithms - the whole raison d'être for Google - was exactly because some content was more deserving than other content.
When I do the search for "maybe there is a god," MeFi doesn't show up until page 8 now. The first two pages of results are virtually all videos and lyrics related to a single 12-year-old Christian pop song.
Which is available for sale on Google Play. Hmm.
The pop song that comprises the first two pages of results for the "maybe there is a god" search is actually called "Mabye There's A Loving God." That's close to the search text, but it's not an exact match.
The MeFi page mentioned in the article contains the exact phrase "maybe there is a god" in the title. And it shows up on page eight of my search results. That is ridiculous.
You can argue that MeFi wasn't singled out on purpose, that algorithms change and there are bigger benefits elsewhere that I'm not seeing in this case and blah blah blah, and I don't know about any of that. No opinion.
But I think you're going to have a hard time arguing that these search results are in any way more accurate, more relevant, or better than the older results were.
And that is very much of a piece with recent, more general criticism of Google's search: That it's simply not as good, not as useful, as it used to be.
Craigslist is a great site too, and you don't see it highly ranked in Google for many queries.
In fact, to me MeFi and HN are about the same, design-wise...
A lot of people have been giving earnest, authoritative sounding SEO advice, but I don't think anyone really knows except Google.
Most people bleat "just write good conteeeeent" and have no idea how Google ranking actually works. I guess what was most galling was seeing certain retail sites getting away with blatant spam while your own sites got mysteriously slammed.
I don't blame Google though - it's a direct result of having a single player market. So go, duckduck, go :)
Commercial sites now have to deal with a full page of ads before the real results so their fate is sealed another way. In other words, we need 4-5 search engines, not one giant monopoly that seeks to increase ad clicks by underhanded methods each quarter.
I bet MeFi could be fixed in a few months if they hired a real SEO expert, ie someone who understands the underlying ranking algos and not just the usual voodoo bullcrap. The site is probably just tripping a few red flags, it could be something as stupid as an accidental link loop in their dynamic content.
I'll give it to Google, once you find your sin and fix it, your traffic will come back, maybe not 100% but a significant recovery. (Unless you have done something really dumb/blackhat like paid links.)
This is not always the case. Look at the Rapgenius event. That was triggered by manual intervention.
>I bet MeFi could be fixed in a few months if they hired a real SEO expert, ie someone who understands the underlying ranking algos
There isn't anyone on the outside that truly understands them because they constantly adjust them to deal with abuse. It is possible that one of these adjustments killed MeFi and it wasn't a recent mistake on MeFi's part.
While they can slap a manual -50 penalty on your domain, from memory you are notified of such penalties in Google Analytics so it's not a silent and nefarious manipulation. MeFi should know if they have such a penalty. I also admit that the "get out of jail" process is even more annoying than dealing with an Apple app store rejection.
So call it manual, algorithmic, display change or whatever, you're f-ed and call moves more of web traffic to its own properties.
Unfortunately, American culture (and many cultures around the world) objectify women and value them based on their looks. So Google's search results may just be a true reflection of what people are interested in looking at. Perhaps Bing's algorithms aren't as good as Google's algorithms in estimating what people might be more interested in.
To be clear, I didn't intend my statement as a feminist critique of society (though personally I would agree with this feminist critique). I just used this example as a roundabout way of defending Google's algorithm and saying that Google gives us the search results we deserve (and ask for) and not the search results that (from Auerback's perspective and mine) we should probably be more interested in.
Perhaps if a different, more specific, standard of measure is applied during query, results will be more applicible for the intended purpose.
For instance substituting accomplished for amazing seems to provide more pointed results.
That made me wonder whether Google has shifted its products from being the kind of thing that their own employee corps would enjoy to things for the general public. I personally f.i. enjoyed the from: to: keywords in Google Maps search. That feature disappeared and probably the majority of users wouldn't use those anyway. Personally I scorn at a search result of the '100 most beautiful women'. But in the large population it will probably lead to a much higher click through. So I wonder if Google Search was oriented towards the smart crowd niche and that they changed focus towards the common and the populous. To put it differently: a shift from an emphasis in information as high value search result towards including entertainment, gossip etc. High brow to low brow.
At the same time, they are trying to do targeted search and I feel like that should be able to help far more than it is.
Sub-optimal, but not terrible.
Removing the quotes gives me similar reaults to you. But there are a few results about interesting women based on achievement not appearance.
Google is trained by the crowd, and the crowd is stupid. See for example the numbers of searches using + operator correctly.
I still use that feature every day. I have a smart keyword setup so that I just type:
into the URL bar and get taken straight to a Google Maps page showing me that route [1]. Forgive me if I've misunderstood you and you're talking about something else, but this feature still exists. AFAIK I'm using the latest version of Google maps.[1] https://www.google.co.uk/maps/preview?q=from+lax+to+san+fran...
The new Google Maps will do multi-point routes [3], but by default when you enter San Francisco to Los Angeles it chooses flights, and hides the option to enter a multi-point route. You have to switch to car directions to get the option to enter more than one point, and enter them one at a time not use the nice from-to-to syntax.
Pretty crap for power users - but I guess ad-supported products make their money by being mass market, and catering to power users isn't very important.
[1] https://maps.google.co.uk/maps?q=from:san+francisco+to:los+a... [2] https://www.google.co.uk/maps/search/from:san+francisco+to:l... [3] https://www.google.co.uk/maps/dir/San+Francisco,+CA,+USA/Los...
It also bugs me that there's no way to enter the mode of transportation as part of the URL query string. I'd love to be able to use smart keyword syntax like:
http://replycam.com/i/MetaFilter_layoffs__Why_has_Google_for...
where is the growth, man? and i don't mean the dropoff. the dropoff is just, you know, change. google changes, all the time, user behaviour changes, user expectation changes, the way google treats sites changes, the internet changes. so yes, 40% less traffic via google organic search happens. you don't have any "right" to google traffic, there are a lot of sites out there.
but the thing that seems to be really broken is the "lack of growth" before the dropoff and after it.
SEO overall has these business cases
as metafilter is a quite popular site with an active community, they should at least do a) and b) quite well, probably doesn't have an issue with c)as there is and was not growth, then there is probably something rotten going on on the platform level.
some issues from a 5 min check i have outlined here https://news.ycombinator.com/item?id=7770414
additionally they seem to completely disregard mobile https://developers.google.com/speed/pagespeed/insights/?url=... and missing out on the probably most important market like ever.
so basically before firing their staff they should have listen to a decent SEO (and implemented some changes) and / or mobile guy.
tl;dr: if your have a popular site with lots and lots of content, but you don't see (organic search) growth then your site is broken, even before google turns of some of your traffic because you know, google.
p.s.: i would love a look into meta-filters google webmaster tools
I actually didn't know that the content MetaFilter has supposedly is good because the design is so off-putting that I've never been able to read much of it.
MetaFilter should really do a redesign to a more readable and friendly interface, that is likely to increase usage/reduce bounce rates (and maybe increase the Google rankings).
I looked it up and it was in fact made in 1999. I doubt anything changed since then.
1. The color combinations are really bad. Dirty-yellow on blue/dirty-yellow on green is close to what a designer would chose if they were planning to make something deliberately look ugly.
2. The titles are in grey while the content text is in white. Why tone down the most important thing on the page?
3. The spacing between all the objects is weird. The space between the comments is huge - it was likely made that large since they would otherwise flow into each other. They should have separated the comments using other methods in addition to just making the space between them huge.
4. For long comments with multiple paragraphs and links it can be hard to see where the next comment begin, since the links (in the comment) and the links to the user profile/permalink to post look almost the same.
Even back in the early 2000's, Metafilter's simple, clean, relatively information-dense design was a welcome change from most news aggregators out there. Metafilter is supposed to be a place for thoughtful medium-to-long-form discussion. I feel that its clean, consistent, text-focused design serves that purpose well.
Honestly, remove comment nesting, move the comment credit line under the comment, and change the color scheme to white and gold on blue, and HN's comment page looks pretty much exactly the same as Metafilter's.
I won't go into the color scheme since I guess that is a matter of taste, but at the same time I don't think it's an accident that you don't see the same color combinations used anywhere else ;)
The comment nesting and placement of the comment credit line(s) is actually really important. It is vital to make the different comments easy to visually separate from each other. HN has one line above and one line below each comment, effectively boxing them in. That, plus the comment nesting, makes HN's comment threads easy to read and skim over.
Metafilter on the other hand has just one line below each comment. Unlike on HN, where the line has a lower contrast/eye focus than the comment text, the line on MF is using the same colors as the links in comments. That is completely counter-productive, and objectively bad design.
Since MF doesn't have comment nesting it should do something else to visually separate the comments. Most other sites subtly change the background color, but there are also other ways to do it.
High contrast/big size should be used for things that should be focused on, and small contrast/small size for things that shouldn't be focused on. On HN the contrast between the color and size of the objects mostly make sense (although self posts should have darker text), but on MF it is all over the place.
I can't help feeling that there's something wrong when the only way to find out why Google has downgraded your site is to attract the attention of one of the core search developers. I get that Google wants to keep it's ranking algorithms secret for fear of black hat SEO poisoning the well even more than it does already. Given Google's dominant position in the search market at the same time it seems wrong to downgrade sites in this fashion with no clarity or visibility over the process whatsoever.
https://news.ycombinator.com/item?id=7763923
It felt like they'd been some sort of tipping point reached and people were just furious with Google for being so opaque.
I've never seen people be so antagonistic with him.
Then again the w/e HN crowd seems to be pretty vicious compared to the week one. (and I honestly feel less intelligent).
Duck Duck Go... perhaps it's time to switch my search engine?
Panda was created to filter down poor content and filter up good content. Sometimes it gets it wrong, but generally if a site is being filtered down it's because there are too many pages of the site in the index that are apparently not helpful to users.
I do think that the design has a lot to do with MetaFilter's demise. I have landed on this site, and, not knowing what I know now after reading articles about the quality information that is on MetaFilter, I have immediately clicked away because it looked like a spammy page. If enough users do this then this shows Google that users don't want to engage with this site.
I haven't had a deep look into the technical structure of the site but there's a good possibility that with a redesign and a review of their architecture, the site could see a return to normal levels with the next Panda refresh.
Any additional insight based on that date?
Full disclosure: MetaFilter member reading about this stuff everywhere with interest who joined here to comment. HN seems right up my alley with the simple design though I'm not sure if I'm ever going to be a threaded comments kind of guy.