The UK Search Awards 2011 will take place on November 3rd at the prestigious Emirates Stadium and will bring together the biggest and brightest stars of the search marketing industry.

A total of 17 categories of awards are to be judged by some of the most experienced names in the search industry.

We’re thrilled to be short-listed for an award for Best use of Search – Retail, for our work with our client BathEmpire. We have some extremely tough competition from a number of agencies much older an larger than us, so even to be short-listed is an honour, and of course the awards night itself promises to be a riot.

I (Nichola, MD of theMediaFlow) do promise to follow-up with photographic evidence of all that transpires on the evening, which will be published on our Facebook page. I’m sure our table will be very well behaved.

I was aware of the Digital Surrey community, and on a recent trip to leading Surrey Online PR Agency, thebluedoor to deliver some advanced SEO training via SEO PR Training; was invited to attend the next event by the team there, who are instrumental in supporting and organising the Digital Surrey events.

The event on April 20th promises to be something extra special, as the format will consist of what is known as Pecha Kucha style, which restricts each speaker to a very strict 6:40 slot.

Although tickets for Digital Surrey April have already “sold” out (it’s free, but you know what I mean) you can register your interest should space become available!

http://digitalsurreyapril.eventbrite.com/

Aaron Wall of SEOBook, a leading SEO resource based in the US, first reported this change in early November for the US market. In his post, Wall posits that for the set of sites that rank well for a term, rather than rank the page that might “traditionally” be the strongest (which is normally any sites’ homepage), Google is then applying “internal site searches & back in other relevancy factors to look for other popular & relevant pages on those sites“.

Looking at our client site rankings, and conferring with industry peers, it seems that this change hit UK shores around December 20th.

Here’s an example:

Simon owns one of the UK’s most established and authoritative sites retailing car parts www.potn.co.uk. The site retails products such as brake discs, alloy wheels, exhausts, high-performance tyres etc. Pretty much anything one may require in order to “pimp one’s ride”. Previously the site homepage has ranked in the top five for almost any term (product) that they retail. Post the 20th December update, Google now return the specific subpage listing for the term in question.

Search “alloy wheels” returns the following result, which was previously the homepage.

Although the listing has slipped a couple of places, one may imagine that the click-through rate on this listing may increase, as the URL shown is descriptive and entirely relevent to the user query. Plus, given the landing page is now the explicit alloy wheels page (on which conversion activity occurs), the revenue effect may be neutral to positive.

What is Happening Here?

Google are taking a further step in the persuit of relevancy and a better user experience, by applying site search (and other relevancy signals) to identify an even more suitable page from within the site; than the naturally more-authoritative homepage.

Who Should this Benefit?

Provided a site is well built, and contains good quality original content, any site can benefit; though in particular this could be useful for ecommerce websites, such as the POTN example above. Ecommerce retailers with multiple products, may now find that the specific product page now ranks where the homepage did previously – which should be great for both user experience and conversions.

Who Are the Losers?

Sites with poor information architecture and little, (or poor quality) content will fail to capitalise on this change. Sites that have failed to invest in content and infrastructure, but have instead chosen to game the algorithm by agressive link-building tactics will also find their poor strategy may no longer pay.

What if Google Are Showing the “Wrong” Page?

In most cases, this change should be quite positive for most quality websites. In some cases however, you may find that a sites blog (by its’ very nature more dynamic and containing more written content than a homepage) usurps a sites’ homepage. We have a client site, with very little written content on the homepage where the blog has usurped the homepage result for the sites core term. In this case, this is not a desirable effect; however we suspect from experience, that the user interaction (as fed back in the search metrics), will prompt the listing to revert to the homepage soon. In the meantime we’re making a couple of on-page changes to the blog, to de-prioritise the site’ core terms.

Future Considerations

This latest change to the Google algorithm makes an even stronger case for the necessity of good quality original content. Site architecture is pushed higher up the agenda. Best-practise in site artchitecture being to establish a flat and wide structure; that is, where content is divided up into as many channels as possible (without comprimising on what makes sense), where each channel is as close as possible to the homepage. That said; if applying site search is indeed the method Google are using to qualify sub-pages, perhaps there is a risk we may see the re-emergence of “SEO landing pages” and low quality keyword-stuffed content again.

Although Bing is currently very much the also-ran, in terms of search engine market share; their partnership with Yahoo! (which will be in effect in the UK before year end) plus their increasing innovations in search technology, plus rather enormous marketing budget means that they will be a serious contender in 2011.

I firmly believe that Bing will start to erode some of Google market share and by the start of 2011 will have at least 10% share if not more. Now is the time to start planning and acting on improving rank on Bing.

What’s the News?

Bing have confirmed that CTR is a factor in their ranking algorithm. This information comes from Barry Schwartz, news editor of Search Engine Land and Schwartz has carefully detailed and personally checked the information with his source, Steve Tullis – Senior Program Manager Lead, Bing. You can read the full story here.

What is CTR?

CTR is the abbreviation for “click-thru rate”. It is a derivative metric determined by dividing the number of times your listing is clicked (clicks), by the total number of times your listing is shown (search impressions), to determine your listing CTR, which is expressed as a percentage.

Example: My site is in position 10 for the term “car insurance”, on Bing, which is searched for 10,000 times a day. I receive 300 clicks a day from Bing, on that listing.

300 /10,000 = 0.03 (expressed as percentage) 3% CTR

What Factors Determine CTR?

Setting aside comparative CTR to competing listings, and the variation afforded by position on-page; in my first-hand experience of working at a search engine, CTR increases or decreases according to assumed relevancy to query term. I.e. the more relevant the page (may be assumed to be by the user,) to the query term, as signified by the title, description and URL of the listing; then the more likely the user will click that listing. Bing generally display the page’ meta title and meta description, as the search result snippet.

What Does this Mean for Rank?

In the Search Engine Land piece, Steve Tullis of Bing is asked if they use CTR as a ranking factor, “yes, we do – but it is one of many factors.” So assuming all other factors determining the ranking order for the sites on a search results page are fairly equal, we might imagine that a listing with compelling, relevant title, description and URL gets a slightly better than average CTR (than might be considered network average,) for the position then this may contribute to that listing moving up the ranking somewhat.

Do I Need to Do Anything?

I’m not a great believer in deliberately optimising for any single known componant of a single search engine’ ranking- algorithm, however in this case we’re talking about click-thru rate; which ranking factors aside, should always be on the agenda for your search engine optimisation team.

I’ve written a rather more detailed piece on four ways to increase organic search CTR for Econsultancy in the past, which is a good place to start. Ultimately, you do need to consider the user journey – in that the query-term, search snippet (normally meta data) and subsequent landing page, all need to provide a relevant and consistent experience. After all; you can optimise the shizzle out of your search result snippet, but if your web page content doesn’t match, you’ve missed an opportunity to convert a potential customer.

I intend to follow up this post with a look at the type of behavioural data CTR as ranking factor provides, plus examine the merits and demerits of CTR as ranking factor.

Confession. I bleed purple. I have an emotional as well as professional motive for attending this session. I’m also becomming a big Bing fan of late and I’m pretty excited to see what they come up with this year.

2009 – Bing has established itself as the “best converting” search engine.

Cedric, 34MM UU in the UK across MSN Windows Live Messenger etc. IT’s not all about reach for you – it’s about performance. Today we’re looking at product capabilities focusing on Bing.

Launched June 2009 and is apparently the “best converting” search engine.

One question we always get. Why bother in search? A – because we are committed to succeed. 5 yrs ago Steve Ballmer said “we want to be number 2”, now they are so how to take it further.

Context – in ten years we have seen 100,000% growth in content. Everything has changed including consumer behaviour. It’s way more difficult to get around the web – hence the rise of search. More and more people are searching more and more.

Queries are becoming more complex in string length and composition. Many more variables to consider and consumers are voicing their frustrations with the search experience. Consumers have evolved faster than search.

Bing Promise: – A brand new search experience. Rich and engaging search content to help British consumers reach more content more quickly.

Intent – we switch on the internet because we want to switch on the internet. Just like TV. Don’t really have an objective in mind.

Bing – capturing  attention using the UI – the attractive imagery. Intended to be a source of information too, with some rollover features. Images are chosen locally by UK journo.

To the search engine nitty gritty.

Many, many queries are purely navigational – therefore why clutter the SERP with anything other than the desired object. E.g. “William Hill” , you would expect the brand result for William Hill.

Informational searches – delivering the answer in the SERP with the feature “instant answer”. E.g. if you type the ref for your flight “BA74″ in the search box then details of the flight will appear in the SERP, giving an instant answer; such as “BA74 departed LAX at 17:45, on time and is due to land at…”.

In the next few weeks expect to see live sports results in the SERP (done in partnership with in-market content providers). Partners include Sky Sports and Toptable.

UX is at the core. Understanding or catering for intent is key. Hence the left navigation. Categorised search for related query term.

Twitter – Bing were 1^st to market with this integration.

Innovation is a core philosophy. Example – using visual stimulus to search, such as searching premiership players by image. Refine by rollover information on the image or refine criteria in the left-hand links. E.g. “Top Scorers” (This looks awesome).

He quotes John Batelle. “Search engine: the database of intent” (I’ll leave it there.)

Over to Adam, who manages the entertainment and gaming channel. We’re gonna get some stats (whohoo).

Reach – 90% of UK internet base.

Unique functionality and tools. MSN targeting capabilities. Adcentre platform has some sophistocated targetting – including days of the week, age and gender.

3^rd year running NNR reports Bing as best converting search engine in the UK.

Bing searchers 47.4% more engaged than Google searchers.

The remainder is an interesting sales pitch for their various AdCenter tools, with UI screenshots. I’ll leave it there. The tool is called “Mate” and has a lot of deep-dive data, demographics, trends, KW data etc. so look it up.

We start off by debunking a few myths. Ralph goes through them; only high PageRank links are good, all paid links are bad etc. Don’t obsess too much about high PR links. PR can go up as well as down.

Another golden rule – not all links are equal. A good link building strategy needs competitive analysis and expertise. Don’t put all your eggs in one basket, as your behaviour needs to look natural. What’s natural about 100 permanent links all a PR5+? Basically; don’t obsess about the “myths” or your behaviour becomes unnatural.

Our MC (conveniently Michael Caselli) asks – Is PR still important?

Ralph – yes. It is indicative of trust. Just don’t obsess and remember the toolbar PR is not the same as the Google internal PR. It’s “for entertainment purposes only”.

Marcus – it can be hard to find sites without paid links so monitoring drops in PR is a good way (when no other factor seems to warrant a penalty), to work out if a source is taking paid links.

Christoph – also be aware of how PR drops can identify linkbuyers. If you have a paid link source and that source drops 2-3PR and you remove the link, you’re identifying yourself to Google – yes, I bought that for the PR. Again Christoph echoes Ralph; don’t obsess about PR and don’t base your strategy on this.

A member of the audience brings up the “bad neighbourhood” term.

Christoph – look at other links on the site linking to you. Are they quality. Are they relevant?

Audience – is there a danger in having links from a site that takes paid links, even if your is not?

Marcus – Yes! (back to bad neighbourhood). Marcus given an analogy “if you’re sat in a bar next to Bob (Rains) and Dave (Naylor) – that’s a bad neighbourhood”. (LOL.)

Christoph – Great links are the ones your competitors can not get easily.

Sebastian – You need to start with analysis, backlink tools, your links, your competitor links. If you believe in white hat only in this industry, then Good Luck! Penalties are over-rated. Too much money in this industry and everyone is buying links. (Basically; they can’t spank the whole industry overnight.)

Ralph goes back to “bad neighbourhood” to give a different perspective – relevancy isn’t too concerning. A bad neighbourhood is more – cheap Viagra, buy Cialis – that sort of stuff. Agrees with Sebastian that this is a highly competitive industry, if you don’t go for black hat, you’re toast.

MC – when is too far?

Ralph – “When you fail” (I love this!)

MC – How many links do I need to get to position one for poker?

Ralph – (I’m paraphrasing) – “How long is a piece of string?” You can have a site with 7000 links in position 1 or a site with 150,000 links in position 1. Remember not all links are equal and your link strategy is just one part of over 200 factors. So many other variables in the mix.

Q from the audience – What is the one “Golden Link”?

Marcus had a great one from W3C. Essentially when they introduced the donation facility on their site, high donors got a link. Marcus was one of the first few – paid his $1k for a year. In his own  words that link was “better than sex”.

Christoph – points out that when such trust is transferred from a link like this, then linking to the next level down the tree is also of great benefit. Being a degree or so removed from that golden source is still worth it. Accumulating such links takes time and also why age/trust are inextricable.

Sebastian – if you develop a new domain, keep it clean, do your press releases. Buy a couple of great paid links; then when the site gets some trust and some traction, then you can really go for it.

Ralph – one of the problems with hunting for that “golden link” is often you can’t tell until after the fact.

MC asks the panel. If you want to rank for the term “Poker”, where is your golden link? Apparently this is a favourite question from Marcus and one he has previously offered money for the right answer for. Marcus thinks this time he could be out of pocket, but it turns out not everybody knows the answer. Which is of course “from the number 1 in the SERP for “poker”. Just ask Google who they trust, and that’s number 1.

Q from the audience – I have a German site and I’m building links from the UK – good or bad?

Sebastian – try to keep a 70/30 rule (70 being the country of origin). Be careful of link brokers here.

Christoph – A link has to pass the smell test. A site in a different country and a different language might make perfect sense; just thing about it logically, naturally and factor relevance.

Q from the audience – How does Google determine relevance? e.g. is “Sportsbook” related to “poker”?

Sebastian – don’t think of Google as a fancy sophisticated bit of .php. Google has been trying for years to emulate human behaviour. Human interpretation of relevance is probably much wider than a machine interpretation.

Ralph – points out that there are linguistic tools available that can assist with this. Also, do bear in mind that text around your anchor text is not ignored. Don’t drop a link with anchor text into a completely irrelevant paragraph.

In summary: Don’t obsess over any one factor in your link-building strategy or you risk losing sight of the whole. Don’t be too timid in this industry or you will sink. Don’t be too rigid or inflexible in your approach or behaviour. Make sure you don’t stink.

If Immanuel Kant was alive today, I think he’d be an SEO on the side. I think the combination of knowledge fuelled by common results of repeated experimentation, combined with the all- too- common Cartesian doubt (did we really drop a place – or did they gain a place) would seriously appeal to him.

So, it is with some sense of compassion that I wanted to reference this hilarious Google result for the search term “Google Ireland”. Have a look at the top ranking results…

Notes on a small Ireland

How good is that?

Maybe someone did a misspell in the meta keywords, cos – like New Zealand is an Ireland, right? Of course the search engines don’t use meta-keywords to inform the ranking algorithm though.

Regardless. My point is that SEO is an inexact science, though we can use learned inferences and experimentation to support our hypotheses, but when all is said and done – even Google can fall foul of their own algorithm.

Diddums.

We can hopefully help you a little further as here we’ve taken the Top 5 Ranking Factors from the SEOmoz report, and explained what they refer to in layman terms.

1. Keyword Focused Anchor Text from External Links

Anchor text refers to the clickable text part of a hyperlink. Anchor text is thought to be weighted quite highly in search engine algorithms as the linked text is normally highly relevant to the landing page. Thus, the anchor text of a link is an indicator of what the linked page is about and can help search engine spiders understand the subject of the target webpage better.

It is best to try and use anchor text which is similar or identical to the keyword targets for your website. Whether the link is from from an internal or external source, even if you can’t use one of your keyword targets, never, ever use a generic phrase such as ‘click here’.

2. External Link Popularity

External link popularity is a measure of the quality and quantity of external links that point to your website. External link popularity is an off-page factor that is supposedly impartial. The theory is that the more links that point to your website, the more popular it is therefore the more useful the pages should be. Content-rich sites should attract lots of links easily; content-poor websites should find it difficult to attract websites.

However it isn’t just about the quantity of links – it’s also about the quality. Not all incoming links are equal; a link from a well-repsected site from a well-respected page (such as the BBC’s homepage) will be worth far, far more than 10 links from an obscure link directory. Naturally, the more respected the website is, the harder it is to get a backlink therefore the more weight the link will carry.

One-way links are thought to be the best kind of link to have (where Site A => Site B only). Reciprocal linking (where Site A => Site B, and Site B => Site A) is useful, particularly if both websites are well respected, however the link is not thought to carry as much weight as a one-way link. Three way links (Site A => Site B => Site C => Site A) are at attempt by some webmasters to create more ‘natural’ looking links. Three way linking can sometimes be better than normal reciprocal links as each link looks like a one-way link.

3. Diversity of Link Sources

Link diversity refers to the number of different root domains that link to your website. Although link quality and link quantity are important, the variety of links is also a critical factor in the search engine algorithms. Example www.themediaflow.com/about and www.themediaflow.com/contact are two different links but are not diverse as the root domain is the same ‘themediaflow’.

The more domains that link to you, the more trust and authority your site is likely to have. Links from a variety of sites are also likely to create new traffic opportunities as well as giving your site exposure to a larger audience. Linking repeatedly from the same domains also looks slightly artificial – a wider sphere of influence means your site will tend to look more authentic.

Having a wide source of links is also a good insurance policy – if you have links from just one or two websites, what if one of the sites loses all of it’s content? You’ve suddenly lost a whole bunch of links. Or what if the search engines adjust their search engine algorithm? You could be out in the cold.

As a final word on link building, when you are building links for your site, don’t just look for the followed links (when links are created, you can apply an HTML attribute called ‘nofollow’ to the link which tells the search engines that the hyperlink should not influence the link target’s ranking in the SERP. Many blogs, directories and bookmarking sites have ‘nofollow’ applied as standard to any link). OK, ‘nofollow’ links won’t pass your website pages any link juice however it is slightly suspicious to only have full-fat links pointing at your website – it will probably look more natural if you have diversity in your links. Also, link building is not just about the SERP – it’s about traffic. Some of the ‘nofollow’ links may actually drive good (free) traffic at your site – and having a diverse source of traffic is great insurance against search engine algorithm changes.

4. Keyword Use Anywhere in the Title Tag

There are two titles that are worthy of mention: the first is within the metadata and the second is within a link.

Metadata title tags should appear in every page on your website and every page on your site should have unique title tags created for it. Titles should always include the keyword targets for that particular page and it is also a useful place to put in alternate spellings in the hope of ranking well (e.g. customisation vs customization). Always front-load your titles with your keywords, but don’t put too many in as this will ‘water down’ the relevance.

Link titles serve two purposes: to help users predict what will happen if they follow a link and to give more relevance pointers to the search engine spiders. Link titles are usually seen by the user when their mouse pointer hovers over a link. The link title should be descriptive, however they should ideally be less than 60 characters and certainly no more than 80 characters long. The link title should be used for supplementary information and to backup the anchor text of a link.

5. Trustworthiness of the Domain Based on Link Distance from Trusted Domains

The link distance refers to how many hops you are away from a ‘Trusted Domain’. The closer you are to a ‘Trusted Domain’ the more trust/authority you inherit from that link.

So what exactly is a ‘Trusted Domain’? Trusted Domains are domains which search engines believe they can trust. Unfortunately, there is no known public list of trusted domains although places such as brands are likely to be on the list. Once a site is classed as a trusted domain, it is believed that any link published within the site will get a little bit of extra link juice. Even if you can’t get a link directly from a trusted domain, by having a link from another domain which does have a link from the trusted domain, you will probably still see a little extra link juice.

I’d say we’re pretty much in agreement with the SEOmoz Top 5 Ranking factors, though there are a number of other factors that can have both a positive and negative effect on your ranking in the SERPs that are also not to be ignored. Finally; one thing that is worth stressing time and again, is that you need to have a fair amount of good- quality, original content that grows and updates frequently. When it comes to SEO you really can not polish a turd.

To illustrate; imagine I’m a small business in the 1960’s seeking to make the public aware of its existence. I’d have to think about physical, visible signage for my premises and advertise in the classifieds. All of that costs money and would involve hiring specialist professionals to create such signage and ad copy. Fast-forward to today, and a small business director has everything they need to make their business brand known; strong and discoverable, just a few keystrokes away.

Of course this works both ways! So, just as a brand owner has some great tools and resources to reach a wider audience more quickly, so does a brand consumer have numerous outlets to transmit their dissatisfaction or voice their opinion. And rightly so; as it’s a good thing this so-called democratisation of the web allows all of us consumers to voice our opinions in an open and public forum. It can be a headache for brands however, as this unregulated, unofficial content on the web vies for attention with your own material. Now; we’re not about to advocate attempting to bury all negative publicity, it is still publicity. What we are advocating is that you seek to ‘own’ your brand message at least on page 1 of the search engines.

Why Own Page 1 for Your Brand Results?

•    Consistency of message
•    Brand recognition and reinforcement
•    Increase traffic to your owned and operated websites
•    Look like a big deal

How to Own Page 1! (It’s easier than you think).

Get Your Domain Name: Make sure you own your brand domain name. www.whatever.com and the local TLD for each market in which you operate; that is if you don’t already! Your own company website is the easiest way to own your first result.
Join Professional Directories: Add yourself to professional and local directories for your sector, thus controlling profile information and having the knock-on benefit of association.
Write and Distribute Press Releases: Writing specific and tailored news about your company is a great way to inform often, and in an official way. Distributing this news as a press release to your online press and having this covered is absolutely invaluable. Trade press publishers naturally tend to rank well due to the authority of their content.
Create a Social Media Presence: Social media is a fantastic, wide-reaching and cost-effective way of essentially, advertising your brand. Facebook pages, Linkedin profiles and CrunchBase profiles are all good ways to extend your brand message. We looked at a couple of good practise tips to be aware of, when expanding your social media brand in a previous post.
Blog and Comment: Nobody knows your business like you do. So get out there and blog, guest blog, comment on related blogs. Establish your internet credentials by showing exactly how much you do know about integrated circuits, or bridal gown design or whatever it is that you do.

Although we can’t control (nor should we want to control) what is said about our brand, but we can control what we say about our brand; so there’s really no excuse not to.

The HTML format allows for the inclusion of a variety of types of metadata, most commonly keywords and description however it can be expanded to highly-granular information such as the Dublin Core and e-GMS standards.

Metadata is typically hidden from a user when they visit a webpage, however a search engine spider will be able to pick up the metadata. Metadata helps a search engine spider understand what a webpage is about, helps to categorise the webpage, and therefore helps to facilitate search and retrieval (NISO, 2001).

Four Key Reasons Why Metadata Matters

1. Meta data works on image-lead pages
It’s a bit of a cliché but there is some truth in the statement oft-used by search engine optimisation (SEO) experts, that “content is King”. Certainly having a lot of content, that changes and updates regularly is a key criteria in optimising for search engines; however there are many valid situations, web pages and entire websites that are very image-lead. As an example many web designer and artist’ websites would naturally contain a gallery of graphical and image lead work. Ensuring you have detailed meta-data describing the site and page content and the nature of your business ensures that search engines have something to get hold of other than just the image content.

2. Meta data expands on existing content
There is often more than one way to describe something – perhaps there is an incorrect way of describing something or there is a commonly used slang expression. If you want to keep your pages corporate and don’t want to include this alternate content however you’d like to rank in the search engine results page (SERP) for the slang, adding them to your metadata may help you rank well for text which doesn’t actually appear on your on-page content.

3. There’s more than one way to spell
There is sometimes more than one way to spell a particular word, for example English spellings tend to prefer the ‘s’ whereas American spellings prefer ‘z’ (e.g. specialisation vs specialization).

If you want to keep your website consistent, you will want to ensure that you keep your descriptions and spelling conventions consistent throughout all your webpages. However what if you run an international website and you want to rank for the alternative spelling?

Metadata can help by allowing you to add these alternate spellings into your metadata. Effectively this data is hidden from your user when they visit the webpage so there are no consistency issues; however it allows search engine spiders to potentially rank your webpage for a variety of other terms.

4. Good metadata increases click-through rates
Search engines will often use the title and meta descriptions of each webpage when they list the result in the SERP. Pages with good metadata are much more likely to be clicked on than ones with poor metadata for example:

Poor metadata – the title of the page is unhelpful, the description doesn’t say anything about the company.

Better metadata – the title of the page explains what the company does and includes keyword targets, the meta description gives more detail about the company’s services and what’s on the page.

If any of the contents of your metadata match up with the phrase the user has searched for, the keyword terms will be bolded up in the SERP. Research has shown that users are more likely to click on a result the more bolded keywords there are in the listing, thus it is important to plan your metadata carefully and ensure that every page on your website has unique, hand-crafted metadata.

However if you leave your metadata out, it’s not the end of the world – search engines will often try and use some of the on-page content in the SERP if you don’t have a meta description and there are “over 200” factors which influence a web page’s position in the SERP for a given keyphrase (that’s the Google secret sauce) therefore you may still rank well even if you leave this key information out. It does tend to be harder and take a lot longer!

References

Google (nd); Google Basics – Serving results
Available from: http://www.google.com/support/webmasters/bin/answer.py?answer=70897&ctx=sibling
[cited: 25 August 2009]

NISO (2001); Understanding Metadata
Available from: http://www.niso.org/publications/press/UnderstandingMetadata.pdf
[cited: 25 August 2009]