Forum Moderators: Robert Charlton & goodroi
In other words, if I use a unique extension, will google discredit the page? Any feedback would be much appreciated.
Thanks!
===================================
Paul
===================================
No, the particular extension used doesn't make any difference to the engines... though, once you've committed to an extension for a page, the engine does associate that extension with a particular url.
So best thing to avoid duplicate content is to pick an extension and stick with it. Thanks for the information!
For one thing, I find a fair number of people manually type in my URLs when they link to me instead of cutting and pasting, and they don't necessarily check their spelling, so I prefer not to throw my users curveballs. IMO there's a lot to be said for keeping things short, simple and obvious.
[edited by: jomaxx at 8:16 pm (utc) on Jan. 3, 2008]
Nonetheless I would use a standard extension such as ".html", not ".bob" as per your example.
I'm kinda partial to "bob"... but with regard to file extensions I do agree. ;)
Choose wisely.
Extensions are so, so, 90s! ;)
I agree with P1R and jomaxx -- extensionless is the way to short, concise, user-friendly, long-lived URLs.
As a matter of fact, the guy who invented (or co-invented) hyperlinks stated as much -- and many years ago: Cool URIs don't change [w3.org].
On a technical note, remember that a URL (or URI) can resolve to any file you like. And Content-type is associated on the server with files. So it's actually not too difficult to point a URL at a file, associate a content-type with its filetype, and have the browser or client correctly identify the content-type -- without any file extension in the URL. Notice also that we speak of URLs on the Web, and then we go throw in this "file type" term -- mixing-up URLs with files, when in fact they are not at all the same thing; URLs locate content on the Web, while filepaths locate content in a server filesystem. There need not be any 'crossover' of these two resource-location systems from the Web side to the filesystem side or vice-versa.
Jim
although the file extension makes no inherent difference to the algorithm, keeping parameters out of the URL does
I recently reorganised a small php/apache ecommerce site (<40 pages) which had previously been mod_rewritten to .htm extensions with keywords in the urls.
The new site uses dynamic urls i.e. product.php?cat_id=1&subcat_id=2 - so dynamic urls with no keywords.
I 301'd all urls individually and took a rankings snapshot before and after then changes. I've monitored weekly for 3 months now. All new urls are in index. Rankings (all top 5 / top 3) for all product pages are COMPLETELY unaffected. The site is in a niche with only a few serious competitors where you would expect that smaller algorithmic advantages/disadvantages would be apparent.
The new site uses dynamic urls i.e. product.php?cat_id=1&subcat_id=2 - so dynamic urls with no keywords.
Why would someone do that? I mean, why would you undo what most are changing to?
Shorter file names, e.g., one word, may also get more respect from Google. They don't look spammy (multiple keywords with hyphenated separations).
I still like the originals: .htm and .html.
p/g
Why would someone do that? I mean, why would you undo what most are changing to?
Because I strongly suspected that it made no difference. As far as my site goes, I was right. If people use your re-written keyword url as a link then you get the anchor benefit, but IN ITSELF it makes absolutely no difference.
Why do unneccessary work?
Has anyone ever done the 'Pepsi challenge' between dynamic and static links in isolation and seen ranking changes? I'd be interested to hear about it. I've read SO much about using static urls for SEO benefit, but as far as I can see it's all just hearsay and 'Chinese whispers'.
Having tried it both ways on one site in a small market where changing a header tag can affect rankings, it made NO difference. In this site I was the person who implemented url rewriting from dynamic to static 3 years ago. It made no difference then either! So a few years down the line when we completely re-worked the db and CMS we scrapped the url rewriting.
The only change I saw is that dynamic urls appear to bleed toolbar PR - all internal PR was lost and remains lost despite the recent update. Site is 4 yrs old.
All rankings rock solid throughout.
I agree with P1R and jomaxx -- extensionless is the way to short, concise, user-friendly, long-lived URLs.
If people use your re-written keyword url as a link then you get the anchor benefit.
Well, there goes one benefit that you no longer have.
But IN ITSELF it makes absolutely no difference.
From my experience it does. It makes a big difference when combined with other elements. There wouldn't be any "IN ITSELF".
Why do unneccessary work?
Because from our perspective, it is necessary. There is much more to it than getting keywords into the URI strings.
Has anyone ever done the 'Pepsi challenge' between dynamic and static links in isolation and seen ranking changes?
Yes, 100's if not 1,000's of us have made the transition. And usually, each time it was for the good. Why? Because there were typically other issues present in the dynamic string that prevented the bot from performing a clean indexing or, even traversing to that content at the end of the string which is usually where the most important stuff sits.
I'd be interested to hear about it. I've read SO much about using static urls for SEO benefit, but as far as I can see it's all just hearsay and 'Chinese whispers'.
Hehehe, Chinese Whispers? I think there will be quite a few who will disagree with you. And, since you have "hard evidence" to the contrary, there really is no need for us to justify now, is there? ;)
Having tried it both ways on one site in a small market where changing a header tag can affect rankings, it made NO difference.
And how was the benchmarking, testing and research done to determine your findings?
All rankings rock solid throughout.
This is not the normal behavior. Rankings are typically affected during the recalculation phase, not sure what you did to bypass those issues.
I have to ask, if rankings were beyond improvement to begin with and remained unchanged, how can you test under those conditions?
Also keep in mind that these days, most of the bots can easily traverse a dynamic URI string. But, there are certain queries that still present issues and those need to be rewritten.
From a user perspective, dynamic URI strings are a nightmare. From a maintenance perspective, they are also a nightmare. Intuitive file naming is a major benefit in today's SEO strategies. It's also an added plus for site administrators.
There is also a Visual Benefit here that hasn't been discussed. :)
What about usability? I don't think extentionless is really "user-friendly" - when I see a link I want to know if it's a regular HTML page or PDF/doc/zip/flash/video/exe/etc file -- and I want to know it before I try to open the link.
You'll only go extensionless for web pages. Documents such as .doc, .pdf, etc. will still have their respective extensions.
[edited by: pageoneresults at 5:25 pm (utc) on Jan. 4, 2008]
You'll only go extensionless for web pages. Documents such as .doc, .pdf, etc. will still have their respective extensions.
domain.com/folder/ vs. domain.com/nonfolder
It would still be confusing - folder versus web page.
Take a look at Google's naming conventions. Also, a look at the W3. Both use extensionless environments. You'll get a 404 if you try adding a trailing forward slash to a resource that has had its extension removed. Or, you may get 301'd back to the correct destination. Depends on how they have it set up.
It makes a big difference when combined with other elements.
how was the benchmarking, testing and research done to determine your findings?
if rankings were beyond improvement to begin with and remained unchanged, how can you test under those conditions?
My whole point is :
if these urls lost some Google 'oomph' from no longer being static keyword urls I would have expected to see drops of at least a place or two - unless it made so little difference as to be negligible. That is what I take from this.
I do take your point about the visual aspects etc though. User experience, call to action in the SERPs etc - good points.
A ranking factor that isn't a ranking factor in its own right?
Very few are.
Its when you get "all of the right ones combined" that counts. And, primary keywords in the URI string are one of them. :)
I checked all rankings, made the change, then checked all rankings after a week, then after all new urls were shown in Google cache.
A week? In Google Cache? That's just the beginning of the process, isn't it? There's still more to come.
Ooops, steering back on topic...
Page Extension - Do the engines care?
I think if they had a choice, the search engines, they'd strip the file extensions too where applicable. I'd bet that the majority of the SE's challenges come in the changing of those extensions. People udating entire platforms and changing the physical address. What a nightmare...
It is the MIME-type in the HTTP header that is important as to how the bot or browser handles the file.
A week? In Google Cache? That's just the beginning of the process, isn't it? There's still more to come.
