It's not frowned upon as long as the words occur after the 3rd slash, ie after the .com/

It's doubtful it helps but it may on trusted big brand sites such as eBay that add user generated titles to their urls which are often spammy.

Trusted sites get away with much more and, right now, that means big corporation backed websites have a lot of wiggle room.

That's kind of what I thought.

So basically what you're saying is it's keyword stuffing if I do it, but since they're a media conglomerate it's a valid tactic which can help them.

You wouldn't recommend it for the average site, I guess is what I'm saying.

Nah. Stuff all you want. What can it hurt? The brands do it all the time! Test it on a few pages (not the site) and see if it makes any difference. These days you can't tell since Panda can't figure out what to do and is run so seldom none of us can figure out if what we did makes any difference, or that what used to work will still work after the next run...

Then again, idiot stuffing deserves what it gets... google isn't the only SE out there and some of the others do have higher standards (bing, etc) which haven't like stuffing anyway.

THERE IS NO MAGIC BULLET.

Don't make your URLs so long they have to be shortened when posted to Twitter.

Nothing wrong with it according to Google. They even tell us to do it.
[static.googleusercontent.com...]

Quadrille's Oft-Quoted 14th Law is quite clear on this issue: More than one hyphen is international shorthand for idiot webmaster; More than two hyphens is Galaxy-wide shorthand for "I'd be a spammer if only I knew how".

But that was written a good while ago, and SEO moves on, indeed, the URL of that article is "/articles/quadrilles-law.shtml".

It is common sense to have a URL that reflects the page content, and most SEOs - and Google - have no problem with intelligent use of hyphens, but the key word there is 'intelligent'.

Folder and file name titles are one of hundreds of factors that Google considers (and, I'd guess, the lesser engines too), and they are very small factors in the great scheme of things.

So while it may be time to update Quadrille's Oft Quoted 14th Law, one part of it needs little revision:

Using hyphens to ensure that a URL reflects page content is is fine, though it has minimal SEO value, but over-use of hyphens is Galaxy-wide shorthand for "I'd be a spammer if only I knew how"

The added advice still applies, too: "Who's counting? Not me. But it's the look of the thing; would you really spend money at:
my-wonderful-domain.info/my-supa-folda/And-anotherfolda/boring-file.com?

It has never been good practice to 'SEO' for irrelevant terms, all that does is raise your bounce rate, and reduce return visitors, and keyword stuffing is as stupid and futile now as it has always been.

And Google isn't impressed, either.

But it's the look of the thing; would you really spend money at:
my-wonderful-domain.info/my-supa-folda/And-anotherfolda/boring-file.com?

Yes, but you may be missing the fact that the vast majority of Internet users just click links. They don't know what a URL is and would not recognise one if they saw it. ;)

Quadrille's Oft-Quoted 14th Law is quite clear on this issue: More than one hyphen is international shorthand for idiot webmaster; More than two hyphens is Galaxy-wide shorthand for "I'd be a spammer if only I knew how".

But that was written a good while ago, and SEO moves on, indeed, the URL of that article is "/articles/quadrilles-law.shtml".

1 !> 1

Yes, but you may be missing the fact that the vast majority of Internet users just click links.
They don't know what a URL is and would not recognise one if they saw it.

Up to a point, that's right; but in Google serps that the spammers obsess about, I've avoided many sad sites on the basis of their spammy title, spammy description or spammy URL. Or, frequently, spammy all three. No point getting in the serps if your site advertises itself as 'one to avoid'!

Many search engine users are naive to spammer tactics; but they know a load of garbage when they see it!

Many search engine users are naive to spammer tactics.

My Mother...

I don't click on those links. Why? Because last time I clicked one, I was infected with some sort of virus thing.

Nothing wrong with it according to Google. They even tell us to do it.

What page in that 32 page document does Google tell us to do this crap?

If your URL contains relevant words, this provides users and search engines with more information about the page than an ID or oddly named parameter would (2).

Are you referring to the above quote from the Google SEO Starter Guide? I don't see that much white space with that recommendation. It doesn't suggest using a long keyword laden URI separated by hyphens or whatever other separators one may use. Heck, lately, I've seen people using curly quotes in their URIs. Damn things break.

Those long keyword hyphenated file names are not friendly at all. Ya, I know, MC does it on his blog. So?!

Let's take a look at the example provided by Shatner. What's the first thing you see wrong with that URI? It exceeds 72 characters. Work with me here, I know, many URIs are longer than 72 characters. But, that doesn't make it friendly. The sample provided gets wrapped in plain text email set to 72 character width. < Ya, people still do that.

http://www.example.com/blue-widgets-green-widgets-sexy-widgets-free-widg
ets-guide-to-widgets-/

It ends up truncating in the middle of widgets. Many times, the </a> is closed at the break so you end up with a portion of the URI not clickable. In many cases there is no truncation algorithm so you end up with individual words being truncated. You know what that means, right? :)

It's even better when the email has been forwarded a few times:

>>>> http://www.example.com/blue-widgets-green-widgets-sexy-widgets-f
> ree-widg
>> ets-guide-to-widgets-/

There is one thing you can design in to the URL structure to help alleviate the problem.

Use http://www.example.com/142892-blue-widgets-green-widgets-sexy-widgets with a record number at the beginning of the path, and where the PHP script checks the slug text is correct for the given ID and then issues a redirect to the correct URL if it is not.

This allows a visitor using any truncated URL, perhaps like http://www.example.com/142892-blue-widgets-gre to still see the content, but crucially, to see it at the correct URL.

Right.

If the link-giver has to say "click on the link or paste it into your browser", you know they've overdone it. Or they're using a badly managed CMS.

Not much point in pasting a broken link into your browser!

Some of our favourite retailers are incapable of a sensible URL - Dindle-Touch-e-Reader-Touch-Screen-Wi-Fi-Special-Offers/dp/B005890G8Y/ref=amb_link_357575542_6?pf_rd_m=ATVPDKIKX0DER&pf_rd_s=gateway-center-column&pf_rd_r=1FEB7HBP8YAWM86GHQS9&pf_rd_t=101&pf_rd_p=1321696362&pf_rd_i=507846 is one of millions of examples. And they still use underscores.

And a hyphen is way preferable to a bunch of 20%s - eg A%2520Touch%2520of%2520the%2520Me

What page in that 32 page document does Google tell us to do this crap?

I see you managed to find it all by yourself. Well done!

Google specifically says ...

URLs with words that are relevant to your site's content and structure are friendlier for visitors navigating your site. Visitors remember them better and might be more willing to link to them.

Google also advises against the use of excessive keywords as in Shatner's original example

avoid using excessive keywords like"baseball-cards-baseball-cards-baseballcards.htm"

Google uses this example ...

www.brandonsbaseballcards.com/articles/ten-rarest-baseball-cards.htm

and states ...

A user performs the query [rarest baseball cards]. One of our deeper pages, its unique description meta tag used as the snippet, appears as a result.

Shatner's example has ten hyphens after the final slash. Google's example has three with no repeated words. Base your conclusions on this.

It's doubtful it helps but it may on trusted big brand sites such as eBay that add user generated titles to their urls which are often spammy.

One of the persons I interviewed recently for an IM position at our company described her previous experience in a small Online marketing company in India where she said her client was eBay and her job profile includes modifying the title, description,etc. on the site in order to increase search engine visibility.

I guess what the OP noticed was something that one of these contract companies did, and not necessarily the big corp.

I suspect that google can handle multi-word urls somewhat ok depending on the context. They know a lot of people are using CMS's like wordpress that can inadvertently spawn large urls, so I don't see them punishing this. However, it does seem clear if your URL has a lot of keyword repetition (like the baseball card example) this is probably bad.

I do think all things being equal the short url will trump the long url. If you have two pages, brandons-old-baseball-cards.html and baseball-cards.html I suspect the latter (all things being equal) would rank better for baseball cards because it's keyword concentration in the URL (my theory) is superior. Google knows your url has to be unique, so I suspect they divide up it's 'ranking power' by its length for starters discourage double-dippers and spammers and secondly a concentrated short url is more likely to be a general on-topic page that google is looking for.

If you do a generic search in google (pick say a bunch of animal) in my experience, the best search results have been short urls.

I got the impression from the google docs as well as other sources that /brandons-dodgers-baseball-cards.html is treated somewhat similarly to a directory separated url like /brandons/dodgers/baseball-cards.html. In my opinion directory structures too should be short and sweet to improve keyword concentration in the URL.

That sounds pretty plausible; to the human eye, an elegant folder structure - with an elegant URL as a direct consequence - is a far more attractive proposition than a spammy mess.

Knowing how Google tries to emulate the human searcher, it is only reasonable that they'd have at least some success in identifying spammy URLs - I'm sure they try.

Of course the corollary is that CMS-generated scrambled eggs - even if not one of 15 duplicates - almost certainly do suffer; not necessarily a 'penalty' of any kind, simply because they are tougher to digest than a mom's-home-cooking simple URL.

That would certainly explain why pages with zero 'link building' and minimal anything else compete so easily with 'professional' cms sites. Or maybe they're just overcooked ;)