These are very interesting links in the context of this class and this week's readings. Both links discuss ways of getting sites properly posted on Google, thereby ushering in new users and Google adminstrators who will approve and support of your site/community showing up in their search results. This is especially relevant since Google attained prominence by catering to smaller sites that normally would have not gotten traffic from the other, big commercial sites (think Yahoo! circa 1999.)
The first link discusses SEOs (search engine optimizers) which webmasters may contract to help their sites come up more often or more prominently in Google searches. This document is very interesting in that Google somewhat advocates and warns of such services. Most interesting in terms of design, is when Google warns of SEOs that offer "shadow" or "doorway" domains that deceptively redirct sites to ones owned/favored by the SEO and pages loaded with many keywords, respetively. Google is adamant that things such as loading a page with thousands of key words will NOT increase a page's chances of being found on their search results.
Which is why the second link is so interesting. In this link focusing on Webmaster, Google gives an essential "How-To" of building and maintaining a site that will be easily found on their search engine. Chief among the advice featured near the bottom of the link: "make pages for users, not search engines," which leads back to the advice about SEOs who offer key word-loaded pages they promise will produce search results.
Google's inherent design theory for a successfull site (where success is defined in how often it pops up in Google's search) is that the more direct and aware of your target audience, the better chance it has of appearing near the top of their results. What's interesting is that Google is offering this theory as a "webmaster guideline" which could debatably be their honest tips for success (such as the long list which i won't paste here starting with "avoid hidden text or links") or their Big Brother-esque warning to "play by our rules and no one gets hurt."
These two pages were pretty straight forward explanations of how to behave and how not to behave when interacting with Google. What I found most interesting from these two readings are the challenges posed by Google's success. They make so much money because they generate huge amounts of traffic through their reputation as the best at search. But as they handle an increasing majority of the searches on the web, there is increasing incentive to manipulate their algorithms, thus making their job more difficult. Furthermore, their success has changed behaviors of web users. (This could have been a reading for last week.)
PageRank is a voting system where the sites with the most votes get to cast 'larger' votes themselves (via links). The assumption of the algorithm seems to be that the most credible sites will receive the most 'votes', which was an accurate assessment of the web in the late '90s. Google's success has provided a new incentive to link (aside from the relevance of content)... profit. I am curious, had e-commerce grown and taken a larger stake in the early web, whether PageRank would have been nearly as successful. Manipulations of the SEOs that Google warns against are attempts to reform the web around the incentive that Google provided. It seems that the incentives generated by PageRank conflict, to some extent, with the the requirements Google sets forth in these documents regarding participation/'membership'.
It actually impresses me that Google has gotten so far on spidering technology that has remained more or less unchanged for a decade. The googlebot inherently believes that everyone's site is a nice organized tree of data files. Realistically, almost everyone now runs "web applications" which means that the concept of a "page" is becoming increasingly antiquated.
From the techincal practictioner perspective, here's where I've been burned in the past by GB:
- No Javascript -- despite the fact that even browsers for small devices now support some level of JS, GB doesn't. So if you have some interface components which are AJAX-y or rendered at runtime in the browser, you will need to do extra work such as having those same links in a block which you immediately make invisible if you want google to catch them.
- No hidden elements -- Do you have "pop up menus" as a navigation tool on your site? That's nice. GB won't follow those.
- Using links for data interface components? -- You'd better make some robots.txt rules for these, because if you have, say, a paginated data display table with links on each column for sorting, AND links to different pages, AND links to next and previous results, GB will gleefully follow every possible combination of these.
- Do pages take time or database resources to render? -- GB doesn't care, it will go ahead and load pages every second or so even if they're taking 10 seconds to render because your already-throttled database is doing unindexed sorts and joins.
- URLs and data elements need to map 1<->1 -- otherwise GB will load and index every possible combination of URL components for that data element (think printable mode, "mobile mode","threaded", and "single message" mode for an online conversation forum.
It turns out that when you're running a website, Google can create quite a bit of work for you -- and since they more or less own online search, it's going to be a while before they have an incentive to "play nice" with dynamic pages.
Submitted by John Blair on Wed, 02/13/2008 - 22:28.
1
point
John Blair
While I'm not the developer that oostendo obviously is, I do understand his post. My only thought while reading the 2 links was, how nice of them, to help us be more Google liked. While I'm a fan of Google for the most part, it seems to me reading oostendo's post and considering the direction Google is taking with their myriad of other non-search related activities, that they may be suffering from letting their core business of search stagnate, thus opening the door for the Google community to look for alternatives - and for competitors to deliver those alternatives.
They have obviously not kept their search technology in pace with web development. What are they going to count on (besides the billions and billions of $$) when/if their other ventures fail? Have they strayed too far from their stated purpose developing areas of practice / expertise outside the core area? Did they ever really have a community to start with, so now they are attempting to start them with these new ventures?
I'm not sure I completely agree with premise of Google not keeping up with cutting edge web development. This problem was faced (and is still faced) when sites use "other means" to convey their message. Adobe Flash sites come to mind. Or sites that rely on images with no ALT tags. These are problems that could adversely affect a site's PageRank. Of course, they also affect usability.
The semantic web seeks to alleviate some of these problems. Google seems to be advocating that webmasters follow certain guidelines which will help improve the ability to index and find information. Obviously, there are ways to make this happen with dynamic sites and these documents seem to be outlining some of these considerations.
There's some stuff you can do to defeat the GB if it's cramping your style.
After creating an account at the Google Webmaster tools (the first step for good SEO), you can limit the rate at which the bot crawls your site. You could also setup your robots.txt to keep yourself safe from the strong metal claws of the spider.
Personally, I love making site maps. By doing so, Google knows where your pages are and knows how often it should come back to them (there's an option to tell google how often the page is refreshed).
Following some not-so-common rules can end up saving you some headache from the Google Bot.
Submitted by LizBlankenship on Wed, 02/13/2008 - 15:13.
4
points
At first I was confused about why these readings were included; although I find them interesting,
what do they have to do with online communities? In light of the other readings for this week, it makes a little more sense. While I may not consider Google Search, a primarily automated, non-social search engine, to be a community, it certainly deals with a lot of the same issues of trust and reputation - the ideas of signalling and screening of Milgram et al. apply perfectly to signaling valuable web content and screening out the good from the bad. Just as more education is costly to achieve but pays off in the end with higher wages and "better" jobs, websites must have better content and relationships with other good sites in order to build reputation and end up with a better ranking, which will make them easier to find for people using Google Search.
Submitted by Matt Adamo on Wed, 02/13/2008 - 17:48.
1
point
When trying to apply this reading to this week's topic, I hadn't even thought of signalling/screening, which of course seem obvious now that Liz mentioned them. I took Google's "Webmaster Guidelines" to be an example of a barrier to entry in Google's index. They're basically a list of things that add up to, "If you want in, this is what you have to do." Google tries to make it pretty clear that there are no shortcuts (i.e. SEOs) around its barriers, as well.
Great application of the readings to signalling and screening. I have been trying to make sense of screening as opposed to signalling, and thought that drawing some very specific examples from these readings would be helpful (for me, and maybe others).
Signaling is an act on the employees part that reveals something to the employer about their preferences. In this reading, unsolicited email messages and guaranteed #1 rank on Google would be signals to webmaster that the SEO is probably not good quality.
Screening is essentially the employers making offers such that employees will have the strongest incentive to accept the offer that reveals true information about themselves. In this reading, Google specifies what offers webmasters can make to potential SEOs to ensure that a bad SEO would not accept the offer, only a good SEO would. This would include specifying in the offer that all intentions must be specified, that your site will not link to the SEO. If the SEO refuses this offer, then they are probably not legit.
Submitted by Satyendra on Thu, 02/14/2008 - 19:56.
2
points
I'm gonig to play devil's advocate here :)
Let's look at the players here:
Assuming Google is the principal:
It would be advantageous for Google to know which sites were good quality and which were bad quality. In fact that's what Google is making billions doing.
Assuming the sites as the agents:
It would be advantageous for the good agents to announce that they are good and differentiate themselves from the bad agents. It would also be advantageous for the bad
agents to pass off as good agents so that they can drive more traffic for their sites.
However, most of the time the 'bad' sites are 'bad' because they want to trick you to reaching their site so they can potentially make some money from you. Therefore since they
are getting some (substaintial?) benefit from you there is a large incentive for them to try to cheat the system, get a higher rank and get you to their site. On the other hand
a large number of good sites .edu, .org pages provide good information but don't have as much of an incentive to get you to their pages. While the Google algorithm has a lot of cool
tricks to prevent the spammers from getting the better of them there isn't really a way the good type can signal their type without the bad type being able to fake it with considerably higher
costs of faking it.
While the Google try to identify the good and the bad type, using link structure, click throughs and other such information using complex algorithms and over time good sites will stand out because of the sites that link to it and raise its page rank I still think there is no provision for a new small good site to declare it is a good site using a signal that is very hard for the bad sites to fake.
I'm not sure why you consider this a devils advocate position? Your interpretation makes sense to me. One thing I would add though is that this is why content is king on the Internet. The hardest thing to fake is having good quality content, in terms of material that is not available elsewhere. So I would argue that in this model it is the content that is signaling to Google. The links and click-throughs will result from good content and then a higher page rank will develop as a consequence.
Submitted by Daniel Zhou on Sat, 02/16/2008 - 01:48.
0
points
Since "signaling" is an effective way to reveal private information, sometimes it is abused. In this case, it is possible that some websites spend the effort in SEO as a signal just to show that their commitment. However, it is totally inappropriate. So Google responses by sending the message that SEO is not a good signal.
I thought this article went a little over the top in downplaying the utility of SEOs. It is clearly in Google's best interest to limit any powerful forces that might challenge the efficiency and accuracy of their alogrithm. This particular FAQ section has a little of a "me thinks they do protest too much" feel to it.
Although I appreciate the transparent tone the article takes, I think that there are clearly some gaps in the information provided. There is a clear agenda put across of deterring site makers from attempting to manipulate search results. AD Words is all about strategically displaying information based on a pay-to-be-displayed-prominently marketing strategy.
I consider Google's Webmaster guidelines useful for any new website.
Any day, it helps to have an organized website, less redundant content.
I'm not suprised that Google has not even given a shred of credibility
to the SEO industry.
The goal is to make it easier for newcomers to find the online community using Google and other search engines.
Design alternative: is change the site's architecture to make it transparent and suitable for Google's crawlers.
I have two additional design suggestions for new web sites:
Sign up for a Google webmaster account to get insight into how well Google is crawling your web pages.
Use a web analytics tool like Google Analytics or other free tools. These tools will provide visitor counts, geographic region, time spent on site, and many more useful statistics that can be used as input while redesigning the site. Spending time and effort on these tools is worth the reward.
I even found a community of practice for the webmasters. Basic search engine optimization is not an arcane craft, but increasing your rank compared to your direct competitors is another ball altogehter.
Google explained their attitudes toward SEO. In Google’s context, it seems that every site in the world is a member of the Google community. If you want your site to be ranked higher through Google search engine, they expect you to follow their rules, including design and content, quality, technical guidelines. I feel Google doesn’t like SEO after reading the links, since some of them just focus on manipulating the algorithms but care noting about the quality of the website. I agree the general guidelines provided by Google, and think a high quality website should deserve a higher ranking. However, somehow I feel their attitude toward SEO is biased, since I found that some SEOs are classified as white hat (http://en.wikipedia.org/wiki/Search_engine_optimization#_note-34), since they do support their clients to use the search engines recommend as part of good design.
In addition, it seems that Google or other search engines include every site into their communities automatically since a site has been launched. For example, it’s true that members (websites) have the right to reject to be crawled by Google, however, they have to follow Google’s rule such as to use robots exclusion standard to prevent their pages to be indexed. My question is that if a site is not asked to become a member of Google, why should the site follow the rule set by Google?
The collection of SEO warnings and advice is essential knowledge for anyone promoting a new community. For new communities targeted to general audiences, search engines are often an economical way to get the word out. SEO strategies are always a part of if not the cornerstone of online marketing campaigns. While most of the Q&A is probably common sense for some of us, any information gap could make someone vulnerable to a deceptive or misguided SEO plan.
Some things to consider when optimizing for search engines:
1) Headlines are really important, especially for blog articles. When your page shows up in search results, a clear and descriptive title is more likely to attract clicks.
2) Google Analytics is your friend. It's a free tool that can be used to better understand how people arrive at your site (i.e. what they searched for), where they browse, and when they leave. It's also a powerful way to measure the effectiveness of conversion pages.
3) People sometimes like ajax and flash, but search engine crawlers don't. Put descriptive text in the alt attributes for images and consider reducing use of ajax and flash.
The Webmaster guidelines for Google search result placement can be broken down into a few core takeaways: don't misrepresent content, don't try to game the system, and do design site architecture with crawlers in mind. Google wants us to believe that if we create high-quality content, people will come and search engine placement will rise, but this feels like a chicken-and-egg problem. Does more traffic raise search engine placement, or does high search engine placement generate more traffic? Or both?
SEO services and best practices for acheiving higher ranks on search engines were new to me. I never realized there was so much behind the search/ranking issues of websites. In SI's Recommender Systems course, PageRank was introduced as one of the recommendation methods that search engines used. In short, Google uses the Pagerank algorith to determine what pages are more important/relevant. Links from one site to another, essentially increases the PageRank of the linked-to site. Basically, a site gets a high PageRank if many sites link to it, and even more so if those link-providing sites have high PageRanks themselves. While this may seem inaccurate, I kind of assumed that high search rankings were simply determined by how well a site was "networked".
Another thing I was confused about was that from what the readings suggest (maybe I am misunderstanding some important concept), Google absolutely has no "favorites" or bias towards displaying certain sites high on its search results. However, doesn't Google search results also differ across users simply because of user preferences? For instance, on default, my google site is www.google.com.tw (chinese). When I input some singer like "Shakira", the first search result I get is www.shakira.com, which is what I would have expected to see. But the 3rd and several top 10 listed sites are Chinese sites. Maybe I am wrong, but somehow I highly doubt that all of these Chinese site have higher PageRank than all the other Shakira-related sites, whether it be in english or another language. Obviously, I'm not saying that I think these sites rigged their way onto the top search results. I'm not sure if this means anything though. Perhaps I miss an important point.
Submitted by Tracy Liu on Thu, 02/14/2008 - 12:16.
0
points
In the aspect of search engine optimization, Google
illustrates his opinion on these optimizers, which in my understanding is
basically "negative". However, I think that this is also a commercial
strategy which strengths Google's reputation on the independent page ranking.
In the aspect of illustrating which kind of site Google
like, it discusses the design and content guideline, technical guidelines and
quality guidelines. I am curious that how Google examine/implement them. For
instance, in the quality control, how Google encourage sites to report another
site which abuses its' quality guidelines, is it highly participated?
what to say, what to say when all of your classmates can decipher the hidden code that is the WEB?????
So, Google dictates search in most Western countries? But it isn't the world's number #1 search engine. So, what standards are these other competing search engines following? Do these other search engines have equal disdain for SOE's (not to be confused with S.O.B.'s)?
For example, Naver is the #1 search engine in South Korea. It tops both Google and YahooSoft...sorry Yahoo. Does anyone know what standards they are using? What keeps them at number one if it is not the magical Google search algorhythm?
My main confusion with this article is how best to frame my thinking of it. Is is a reference to how websites can create barriers to entry to be a part of them, as Google does by punishing bad SEOs and sites that hire them? Or is it in fact an example of signaling and screening, in that there are self-selection constraints evident that mean that it's in a sites best interest to follow the rules and show its credibility as a site to be indexed?
I'm more inclined to think of it as the former - that sites like Google use these strongly worded suggestions to create important barriers to keep up the quality of its 'community'. These are different from terms of service - Google is not saying "you cannot work with a SEO", but they are providing these as guidelines to show how it is in your best interest to know how Google operates and how to follow the rules.
- Contain anything designed to mislead, confuse, or defraud the user in any way.
- Put links into feed stories and notifications that trick users into installing another application.
- Track visits to a user’s profile, whether aggregated anonymously or identified individually.
- Promote other applications in notifications in order to pool
notifications together and work around the limit of 20 notifications
per user per application each day.
There is also information on what is allowable, which discussing topics such as what forms of advertising are allowed. So, overall, I'm not sure how many people are into Facebook apps, but this seems like a great corollary to the Google guidelines - with sites like Google and Facebook that have gotten large and popular enough, there will be people who try to game their system to benefit themselves. So, it becomes necessary to create barriers and rules against these behaviors, and reinforce what actions are preferred instead.
As a webmaster for the Central Bank of Malaysia, I tried to follow these advisory pages rather religiously -- even created Sitemaps and robots.txt.. But initially I wasn't able to get my organization to be on top, whatever the search term - bnm, bank negara, central bank.. however, over the years, my site's reputation improved and other sites started linking to it, and walla.. it's now always #1..
Of course in the in-between time, many companies approached us to sell theire SEO strategy, which I didn't buy.. to me, reputation is key.. you may not be on top quickly, but once u get there, you'll stick there..
This piece can be seen as a Google's regulation rules for players in
the huge community connected by Google's search service, whereby
people/agents can interact with each other. For these two guidelines,
although both for making the mechanism more efficient in general, one
is dressing the benefit for normal users who want to be found by others
through Google, and the other, intends to enhance the role of Google
playing to prohibit abuse.
Then we can see there are different
roles (naturally, not assigned by the system administrator) people
playing in a dynamic community. For the website masters and the SEO
companies, there is always the incentive to abuse the Google rule for
their interest. In addition, there is always controversy where is the
line between good or unethetical. It is always the challenge for a
community administration.
I am particularly curious about why Google
likes to even emphasize on these kinds of possible abuse happening
rather than avoiding mentioning. It is really a cultural difference
thing as my opinion. If a Chinese thing, I guess the administrator
would like to pretend no such things and prohibit it by a more implicit
way. Their logic is that this claim would just make people who did not
know about it but can actually learn from it. It is just like the
metaphor that people often learned how to rob from Hollywood movies.
The software development is always behind the hackers. Google 's
ranking algorithm is particularly suffering from this kind of abuse,
since it had the assumption which is not really true. It is based on
something the agents who are be ranked can actually manipulate. In
fact, many Chinese websites are extremely active in doing this...on
Google, a website has a lot of irrelevant keywords but nothing actual
information there, which would be a real problem Google is facing.
It is interesting that Google even has these pages about SEO. If you think about, they are in no way obligated to discuss this, and it is not even clear what kind of incentive(s) they may have to do it.
One thing that strikes me is how archaic the strategies are: optimizing your webpages for Lynx? It makes me wonder what the barriers are to a truly useful WWW which is able to index and search images, audio, dynamic content, etc.
One key (political/economic) problem this raises is that of search standards. The search ecology would function much better if a diverse collection of search sites used an XML-ish (or some other) standard for indexing and sharing information, and if those standards were open and publicly available with rules for SEO related to the standards AND NOT to the particular implementation of any given proprietary search engine. More to the point, a good and open set of search and SEO standards would mean that anyone could write a search website or search application that could transmit/receive queries. This would result in much more rapid innovation in the "search" economy.
-------------------------------------------------------- PHartzog@umich.edu
--------------------------------------------------------
The Universe is made up of stories, not atoms.
--Muriel Rukeyser
It seems that the attempt to manipulate Google’ pagerank is growing all the time. In the first answer page, Google seems to try to explain and inform some of its users not to trust those potential manipulators as well as a warning to them. In the second one, Google tries to provide some guidelines that it hopes most website managers could follow. While reminds may be useful to less knowledgeable users, it is less likely to prevent all manipulations as long as the benefits it will bring outweigh the costs bearing. As we have come across in last week’s readings, laws and rules are really important in regulating users’ behaviors, however, it seems that such rules are just too difficult to implement in this case.
Google has become the virtual gateway for Internet. Considering how
much our lives depend on Internet, is it fair to assume that one
"commercial" company controls the gateway to the "free world"?
The post clearly defines what you can do and what you should not do
for SEO. Unfortunately, these standards (although designed specifically
for Google) have become the de-facto standards for anyone publishing a
webpage. Afterall who wants to be black-listed from the Google.
Should there be open standards for SEO which search engines as well
as the SEO companies should follow? We can have a governing body
(ofcourse Google will be a part of it) to create standards. But atleast
we will not see cases in which Google blocks indexing the website for
one reason or the other.
Opening the community to Google
These are very interesting links in the context of this class and this week's readings. Both links discuss ways of getting sites properly posted on Google, thereby ushering in new users and Google adminstrators who will approve and support of your site/community showing up in their search results. This is especially relevant since Google attained prominence by catering to smaller sites that normally would have not gotten traffic from the other, big commercial sites (think Yahoo! circa 1999.)
The first link discusses SEOs (search engine optimizers) which webmasters may contract to help their sites come up more often or more prominently in Google searches. This document is very interesting in that Google somewhat advocates and warns of such services. Most interesting in terms of design, is when Google warns of SEOs that offer "shadow" or "doorway" domains that deceptively redirct sites to ones owned/favored by the SEO and pages loaded with many keywords, respetively. Google is adamant that things such as loading a page with thousands of key words will NOT increase a page's chances of being found on their search results.
Which is why the second link is so interesting. In this link focusing on Webmaster, Google gives an essential "How-To" of building and maintaining a site that will be easily found on their search engine. Chief among the advice featured near the bottom of the link: "make pages for users, not search engines," which leads back to the advice about SEOs who offer key word-loaded pages they promise will produce search results.
Google's inherent design theory for a successfull site (where success is defined in how often it pops up in Google's search) is that the more direct and aware of your target audience, the better chance it has of appearing near the top of their results. What's interesting is that Google is offering this theory as a "webmaster guideline" which could debatably be their honest tips for success (such as the long list which i won't paste here starting with "avoid hidden text or links") or their Big Brother-esque warning to "play by our rules and no one gets hurt."
Mo' Money Mo' Problems
These two pages were pretty straight forward explanations of how to behave and how not to behave when interacting with Google. What I found most interesting from these two readings are the challenges posed by Google's success. They make so much money because they generate huge amounts of traffic through their reputation as the best at search. But as they handle an increasing majority of the searches on the web, there is increasing incentive to manipulate their algorithms, thus making their job more difficult. Furthermore, their success has changed behaviors of web users. (This could have been a reading for last week.)
PageRank is a voting system where the sites with the most votes get to cast 'larger' votes themselves (via links). The assumption of the algorithm seems to be that the most credible sites will receive the most 'votes', which was an accurate assessment of the web in the late '90s. Google's success has provided a new incentive to link (aside from the relevance of content)... profit. I am curious, had e-commerce grown and taken a larger stake in the early web, whether PageRank would have been nearly as successful. Manipulations of the SEOs that Google warns against are attempts to reform the web around the incentive that Google provided. It seems that the incentives generated by PageRank conflict, to some extent, with the the requirements Google sets forth in these documents regarding participation/'membership'.
GoogleBot Woes for Web Applications
It actually impresses me that Google has gotten so far on spidering technology that has remained more or less unchanged for a decade. The googlebot inherently believes that everyone's site is a nice organized tree of data files. Realistically, almost everyone now runs "web applications" which means that the concept of a "page" is becoming increasingly antiquated.
From the techincal practictioner perspective, here's where I've been burned in the past by GB:
- No Javascript -- despite the fact that even browsers for small devices now support some level of JS, GB doesn't. So if you have some interface components which are AJAX-y or rendered at runtime in the browser, you will need to do extra work such as having those same links in a block which you immediately make invisible if you want google to catch them.
- No hidden elements -- Do you have "pop up menus" as a navigation tool on your site? That's nice. GB won't follow those.
- Using links for data interface components? -- You'd better make some robots.txt rules for these, because if you have, say, a paginated data display table with links on each column for sorting, AND links to different pages, AND links to next and previous results, GB will gleefully follow every possible combination of these.
- Do pages take time or database resources to render? -- GB doesn't care, it will go ahead and load pages every second or so even if they're taking 10 seconds to render because your already-throttled database is doing unindexed sorts and joins.
- URLs and data elements need to map 1<->1 -- otherwise GB will load and index every possible combination of URL components for that data element (think printable mode, "mobile mode","threaded", and "single message" mode for an online conversation forum.
It turns out that when you're running a website, Google can create quite a bit of work for you -- and since they more or less own online search, it's going to be a while before they have an incentive to "play nice" with dynamic pages.
--------------------------------------------------------------
oostendo@umich.edu
--------------------------------------------------------------
In Google we trust
John Blair
While I'm not the developer that oostendo obviously is, I do understand his post. My only thought while reading the 2 links was, how nice of them, to help us be more Google liked. While I'm a fan of Google for the most part, it seems to me reading oostendo's post and considering the direction Google is taking with their myriad of other non-search related activities, that they may be suffering from letting their core business of search stagnate, thus opening the door for the Google community to look for alternatives - and for competitors to deliver those alternatives.
They have obviously not kept their search technology in pace with web development. What are they going to count on (besides the billions and billions of $$) when/if their other ventures fail? Have they strayed too far from their stated purpose developing areas of practice / expertise outside the core area? Did they ever really have a community to start with, so now they are attempting to start them with these new ventures?
Advocating the semantic web
I'm not sure I completely agree with premise of Google not keeping up with cutting edge web development. This problem was faced (and is still faced) when sites use "other means" to convey their message. Adobe Flash sites come to mind. Or sites that rely on images with no ALT tags. These are problems that could adversely affect a site's PageRank. Of course, they also affect usability.
The semantic web seeks to alleviate some of these problems. Google seems to be advocating that webmasters follow certain guidelines which will help improve the ability to index and find information. Obviously, there are ways to make this happen with dynamic sites and these documents seem to be outlining some of these considerations.
Us Vs. GB
There's some stuff you can do to defeat the GB if it's cramping your style.
After creating an account at the Google Webmaster tools (the first step for good SEO), you can limit the rate at which the bot crawls your site. You could also setup your robots.txt to keep yourself safe from the strong metal claws of the spider.
Personally, I love making site maps. By doing so, Google knows where your pages are and knows how often it should come back to them (there's an option to tell google how often the page is refreshed).
Following some not-so-common rules can end up saving you some headache from the Google Bot.
relating to the other readings
At first I was confused about why these readings were included; although I find them interesting,
what do they have to do with online communities? In light of the other readings for this week, it makes a little more sense. While I may not consider Google Search, a primarily automated, non-social search engine, to be a community, it certainly deals with a lot of the same issues of trust and reputation - the ideas of signalling and screening of Milgram et al. apply perfectly to signaling valuable web content and screening out the good from the bad. Just as more education is costly to achieve but pays off in the end with higher wages and "better" jobs, websites must have better content and relationships with other good sites in order to build reputation and end up with a better ranking, which will make them easier to find for people using Google Search.
Barriers to entry
When trying to apply this reading to this week's topic, I hadn't even thought of signalling/screening, which of course seem obvious now that Liz mentioned them. I took Google's "Webmaster Guidelines" to be an example of a barrier to entry in Google's index. They're basically a list of things that add up to, "If you want in, this is what you have to do." Google tries to make it pretty clear that there are no shortcuts (i.e. SEOs) around its barriers, as well.
Specific Application to Signalling & Screening
Great application of the readings to signalling and screening. I have been trying to make sense of screening as opposed to signalling, and thought that drawing some very specific examples from these readings would be helpful (for me, and maybe others).
Signaling is an act on the employees part that reveals something to the employer about their preferences. In this reading, unsolicited email messages and guaranteed #1 rank on Google would be signals to webmaster that the SEO is probably not good quality.
Screening is essentially the employers making offers such that employees will have the strongest incentive to accept the offer that reveals true information about themselves. In this reading, Google specifies what offers webmasters can make to potential SEOs to ensure that a bad SEO would not accept the offer, only a good SEO would. This would include specifying in the offer that all intentions must be specified, that your site will not link to the SEO. If the SEO refuses this offer, then they are probably not legit.
The need for a good signal
I'm gonig to play devil's advocate here :)
Let's look at the players here:
Assuming Google is the principal:
It would be advantageous for Google to know which sites were good quality and which were bad quality. In fact that's what Google is making billions doing.
Assuming the sites as the agents:
It would be advantageous for the good agents to announce that they are good and differentiate themselves from the bad agents. It would also be advantageous for the bad
agents to pass off as good agents so that they can drive more traffic for their sites.
However, most of the time the 'bad' sites are 'bad' because they want to trick you to reaching their site so they can potentially make some money from you. Therefore since they
are getting some (substaintial?) benefit from you there is a large incentive for them to try to cheat the system, get a higher rank and get you to their site. On the other hand
a large number of good sites .edu, .org pages provide good information but don't have as much of an incentive to get you to their pages. While the Google algorithm has a lot of cool
tricks to prevent the spammers from getting the better of them there isn't really a way the good type can signal their type without the bad type being able to fake it with considerably higher
costs of faking it.
While the Google try to identify the good and the bad type, using link structure, click throughs and other such information using complex algorithms and over time good sites will stand out because of the sites that link to it and raise its page rank I still think there is no provision for a new small good site to declare it is a good site using a signal that is very hard for the bad sites to fake.
Devil's advocate?
I'm not sure why you consider this a devils advocate position? Your interpretation makes sense to me. One thing I would add though is that this is why content is king on the Internet. The hardest thing to fake is having good quality content, in terms of material that is not available elsewhere. So I would argue that in this model it is the content that is signaling to Google. The links and click-throughs will result from good content and then a higher page rank will develop as a consequence.
don't abuse signaling
Since "signaling" is an effective way to reveal private information, sometimes it is abused. In this case, it is possible that some websites spend the effort in SEO as a signal just to show that their commitment. However, it is totally inappropriate. So Google responses by sending the message that SEO is not a good signal.
SEO Blacklisting
I thought this article went a little over the top in downplaying the utility of SEOs. It is clearly in Google's best interest to limit any powerful forces that might challenge the efficiency and accuracy of their alogrithm. This particular FAQ section has a little of a "me thinks they do protest too much" feel to it.
Although I appreciate the transparent tone the article takes, I think that there are clearly some gaps in the information provided. There is a clear agenda put across of deterring site makers from attempting to manipulate search results. AD Words is all about strategically displaying information based on a pay-to-be-displayed-prominently marketing strategy.
Lisa McLaughlin
Use Analytics and Webmaster tools as well
I consider Google's Webmaster guidelines useful for any new website.
Any day, it helps to have an organized website, less redundant content.
I'm not suprised that Google has not even given a shred of credibility
to the SEO industry.
The goal is to make it easier for newcomers to find the online community using Google and other search engines.
Design alternative: is change the site's architecture to make it transparent and suitable for Google's crawlers.
I have two additional design suggestions for new web sites:
I even found a community of practice for the webmasters. Basic search engine optimization is not an arcane craft, but increasing your rank compared to your direct competitors is another ball altogehter.
Some Thoughts
Google explained their attitudes toward SEO. In Google’s context, it seems that every site in the world is a member of the Google community. If you want your site to be ranked higher through Google search engine, they expect you to follow their rules, including design and content, quality, technical guidelines. I feel Google doesn’t like SEO after reading the links, since some of them just focus on manipulating the algorithms but care noting about the quality of the website. I agree the general guidelines provided by Google, and think a high quality website should deserve a higher ranking. However, somehow I feel their attitude toward SEO is biased, since I found that some SEOs are classified as white hat (http://en.wikipedia.org/wiki/Search_engine_optimization#_note-34), since they do support their clients to use the search engines recommend as part of good design.
In addition, it seems that Google or other search engines include every site into their communities automatically since a site has been launched. For example, it’s true that members (websites) have the right to reject to be crawled by Google, however, they have to follow Google’s rule such as to use robots exclusion standard to prevent their pages to be indexed. My question is that if a site is not asked to become a member of Google, why should the site follow the rule set by Google?
SEO, Design Guidelines
The collection of SEO warnings and advice is essential knowledge for anyone promoting a new community. For new communities targeted to general audiences, search engines are often an economical way to get the word out. SEO strategies are always a part of if not the cornerstone of online marketing campaigns. While most of the Q&A is probably common sense for some of us, any information gap could make someone vulnerable to a deceptive or misguided SEO plan.
Some things to consider when optimizing for search engines:
1) Headlines are really important, especially for blog articles. When your page shows up in search results, a clear and descriptive title is more likely to attract clicks.
2) Google Analytics is your friend. It's a free tool that can be used to better understand how people arrive at your site (i.e. what they searched for), where they browse, and when they leave. It's also a powerful way to measure the effectiveness of conversion pages.
3) People sometimes like ajax and flash, but search engine crawlers don't. Put descriptive text in the alt attributes for images and consider reducing use of ajax and flash.
More SEO tools here: http://techmagazine.ws/the-best-seo-tools/
The Webmaster guidelines for Google search result placement can be broken down into a few core takeaways: don't misrepresent content, don't try to game the system, and do design site architecture with crawlers in mind. Google wants us to believe that if we create high-quality content, people will come and search engine placement will rise, but this feels like a chicken-and-egg problem. Does more traffic raise search engine placement, or does high search engine placement generate more traffic? Or both?
Confused about how PageRank & Google search results work
SEO services and best practices for acheiving higher ranks on search engines were new to me. I never realized there was so much behind the search/ranking issues of websites. In SI's Recommender Systems course, PageRank was introduced as one of the recommendation methods that search engines used. In short, Google uses the Pagerank algorith to determine what pages are more important/relevant. Links from one site to another, essentially increases the PageRank of the linked-to site. Basically, a site gets a high PageRank if many sites link to it, and even more so if those link-providing sites have high PageRanks themselves. While this may seem inaccurate, I kind of assumed that high search rankings were simply determined by how well a site was "networked".
Another thing I was confused about was that from what the readings suggest (maybe I am misunderstanding some important concept), Google absolutely has no "favorites" or bias towards displaying certain sites high on its search results. However, doesn't Google search results also differ across users simply because of user preferences? For instance, on default, my google site is www.google.com.tw (chinese). When I input some singer like "Shakira", the first search result I get is www.shakira.com, which is what I would have expected to see. But the 3rd and several top 10 listed sites are Chinese sites. Maybe I am wrong, but somehow I highly doubt that all of these Chinese site have higher PageRank than all the other Shakira-related sites, whether it be in english or another language. Obviously, I'm not saying that I think these sites rigged their way onto the top search results. I'm not sure if this means anything though. Perhaps I miss an important point.
Commercial strategy? Incentive to report other site?
In the aspect of search engine optimization, Google
illustrates his opinion on these optimizers, which in my understanding is
basically "negative". However, I think that this is also a commercial
strategy which strengths Google's reputation on the independent page ranking.
In the aspect of illustrating which kind of site Google
like, it discusses the design and content guideline, technical guidelines and
quality guidelines. I am curious that how Google examine/implement them. For
instance, in the quality control, how Google encourage sites to report another
site which abuses its' quality guidelines, is it highly participated?
Never,Naver,Never,Naver
what to say, what to say when all of your classmates can decipher the hidden code that is the WEB?????
So, Google dictates search in most Western countries? But it isn't the world's number #1 search engine. So, what standards are these other competing search engines following? Do these other search engines have equal disdain for SOE's (not to be confused with S.O.B.'s)?
For example, Naver is the #1 search engine in South Korea. It tops both Google and YahooSoft...sorry Yahoo. Does anyone know what standards they are using? What keeps them at number one if it is not the magical Google search algorhythm?
http://www.naver.com/
How to frame the discussion of this article?
My main confusion with this article is how best to frame my thinking of it. Is is a reference to how websites can create barriers to entry to be a part of them, as Google does by punishing bad SEOs and sites that hire them? Or is it in fact an example of signaling and screening, in that there are self-selection constraints evident that mean that it's in a sites best interest to follow the rules and show its credibility as a site to be indexed?
I'm more inclined to think of it as the former - that sites like Google use these strongly worded suggestions to create important barriers to keep up the quality of its 'community'. These are different from terms of service - Google is not saying "you cannot work with a SEO", but they are providing these as guidelines to show how it is in your best interest to know how Google operates and how to follow the rules.
In a similar vein, I found an article on how Facebook now has guidelines and suggestions for application developers, which clearly explain their policies and how to keep from getting punished by Facebook for breaking them.
Among the DO NOT's:
- Contain anything designed to mislead, confuse, or defraud the user in any way.
- Put links into feed stories and notifications that trick users into installing another application.
- Track visits to a user’s profile, whether aggregated anonymously or identified individually.
- Promote other applications in notifications in order to pool
notifications together and work around the limit of 20 notifications
per user per application each day.
There is also information on what is allowable, which discussing topics such as what forms of advertising are allowed. So, overall, I'm not sure how many people are into Facebook apps, but this seems like a great corollary to the Google guidelines - with sites like Google and Facebook that have gotten large and popular enough, there will be people who try to game their system to benefit themselves. So, it becomes necessary to create barriers and rules against these behaviors, and reinforce what actions are preferred instead.
going up the pagerank
As a webmaster for the Central Bank of Malaysia, I tried to follow these advisory pages rather religiously -- even created Sitemaps and robots.txt.. But initially I wasn't able to get my organization to be on top, whatever the search term - bnm, bank negara, central bank.. however, over the years, my site's reputation improved and other sites started linking to it, and walla.. it's now always #1..
Of course in the in-between time, many companies approached us to sell theire SEO strategy, which I didn't buy.. to me, reputation is key.. you may not be on top quickly, but once u get there, you'll stick there..
The Giant Community, which is even hard to regulate
This piece can be seen as a Google's regulation rules for players in
the huge community connected by Google's search service, whereby
people/agents can interact with each other. For these two guidelines,
although both for making the mechanism more efficient in general, one
is dressing the benefit for normal users who want to be found by others
through Google, and the other, intends to enhance the role of Google
playing to prohibit abuse.
Then we can see there are different
roles (naturally, not assigned by the system administrator) people
playing in a dynamic community. For the website masters and the SEO
companies, there is always the incentive to abuse the Google rule for
their interest. In addition, there is always controversy where is the
line between good or unethetical. It is always the challenge for a
community administration.
I am particularly curious about why Google
likes to even emphasize on these kinds of possible abuse happening
rather than avoiding mentioning. It is really a cultural difference
thing as my opinion. If a Chinese thing, I guess the administrator
would like to pretend no such things and prohibit it by a more implicit
way. Their logic is that this claim would just make people who did not
know about it but can actually learn from it. It is just like the
metaphor that people often learned how to rob from Hollywood movies.
The software development is always behind the hackers. Google 's
ranking algorithm is particularly suffering from this kind of abuse,
since it had the assumption which is not really true. It is based on
something the agents who are be ranked can actually manipulate. In
fact, many Chinese websites are extremely active in doing this...on
Google, a website has a lot of irrelevant keywords but nothing actual
information there, which would be a real problem Google is facing.
SEO
It is interesting that Google even has these pages about SEO. If you think about, they are in no way obligated to discuss this, and it is not even clear what kind of incentive(s) they may have to do it.
One thing that strikes me is how archaic the strategies are: optimizing your webpages for Lynx? It makes me wonder what the barriers are to a truly useful WWW which is able to index and search images, audio, dynamic content, etc.
One key (political/economic) problem this raises is that of search standards. The search ecology would function much better if a diverse collection of search sites used an XML-ish (or some other) standard for indexing and sharing information, and if those standards were open and publicly available with rules for SEO related to the standards AND NOT to the particular implementation of any given proprietary search engine. More to the point, a good and open set of search and SEO standards would mean that anyone could write a search website or search application that could transmit/receive queries. This would result in much more rapid innovation in the "search" economy.
--------------------------------------------------------
PHartzog@umich.edu
--------------------------------------------------------
The Universe is made up of stories, not atoms.
--Muriel Rukeyser
It seems that the attempt to
It seems that the attempt to manipulate Google’ pagerank is growing all the time. In the first answer page, Google seems to try to explain and inform some of its users not to trust those potential manipulators as well as a warning to them. In the second one, Google tries to provide some guidelines that it hopes most website managers could follow. While reminds may be useful to less knowledgeable users, it is less likely to prevent all manipulations as long as the benefits it will bring outweigh the costs bearing. As we have come across in last week’s readings, laws and rules are really important in regulating users’ behaviors, however, it seems that such rules are just too difficult to implement in this case.
Open Standards for SEO : Revised posting from blog
Google has become the virtual gateway for Internet. Considering how
much our lives depend on Internet, is it fair to assume that one
"commercial" company controls the gateway to the "free world"?
The post clearly defines what you can do and what you should not do
for SEO. Unfortunately, these standards (although designed specifically
for Google) have become the de-facto standards for anyone publishing a
webpage. Afterall who wants to be black-listed from the Google.
Should there be open standards for SEO which search engines as well
as the SEO companies should follow? We can have a governing body
(ofcourse Google will be a part of it) to create standards. But atleast
we will not see cases in which Google blocks indexing the website for
one reason or the other.