Forum Moderators: open
<form method=get>
to
<form method=post>
On the tech end, the above switch would mean that referral strings from search engines would be a thing of the past. No more keyword work, no more knowing which kw's people were using to find your site, no more checking your rankings. The net effect would be the end of seo.
The above switch would not be without merit or precedent. Both AOL and Overture routinely encode urls and session strings. By moving to POSTED data, they would eliminate some of the need for that.
On the down side, the se's would have to come up with new logging tools for themselves. Like us, they use referral strings to trace visitor footsteps through their own site. For them, it is a great way to study how people search. They can analyze "blue widgets" become "blue fuzzy widgets" and gauge surfer search success rates. If a switch is made, they would need a new set of logging tools to study that.
There are some meta engines and java script based forms on engines that already hide referral strings. Most of us have seen this effect from search MSN already.
I guess we've know the engines could do the above switch for years. It was always in their best interest to leave things as a GET. That is especially true for engines that charge fees for listings. Yahoo, Inktomi, and Overture really need those referral strings present for their customers sites - they need to see those keywords.
What effect do you think it would have on SEO if some popular engine made the switch?
Will browsers allow people to surf without being prompted to resubmit the form? Opera already does and most of the mobile browsers do too.
You would still have the data regarding the URL so would know SE 'alpha' had sent you n referrals.
You would not have the keyword data to analyse. Thus the analysis and improvement of the site would be more fun (difficult)
Keyword tools are widely available for basic research and "seat of the pants" can get you pretty far down the road, I'm not so sure it would be an SEO deterrent, it would probably just spawn more "voodoo" practitioners who prey on gullible clients.
Once the set of header information is captured from the forms, it can be reproduced by any utility or tool. They can generate the appropriate headers with hidden fields and values set.
SEs could have denied URLs where the source IP didn't match their own (and some do), so the above would work as well, as long this non-blocking continues.
Now for referers... Well, that is a bummer. Unless the SEs cooperate, the information would be lost.
I do feel bad that all of the witch doctors, er I mean seo's will be out of work. :)
When you phone a company for any kind of services, in response to an advert you've seen somewhere, one of the first questions the business wants to know is: "...where did you see our advert?"
It tells them which adverts worked and where they work best.
Perhaps the SEs will offer this as an in-house service (you subscribe to their stats for a fee).
It's a bit more complicated, but it can be done. If it is put in place then I would see a bit of a shake out in the less technically adept seo-ers, but I would see this as removing most of the cowboys...
Interesting times ahead still!
Its like 3 different people taking three diferent routes to a party. Then the host complaining that he cant throw a good party cuz he doesnt know how they got to his house. It doesnt make a lot of sense.
How many webmasters arrived at some engine, say Google, Alltheweb, or Altavista, even Yahoo!, because they got a referral? And they could see the words that were used to get to their page?
Without that kind of subtle advertising, how many engines today would be where they are? I doubt very many.
My motivation for paying anything, to any search engine, would be gone - completely. If they let me have those referral strings, maybe I'd pay, perhaps I wouldn't - but without that information, there is no need that I can see to pay somebody for spidering or inclusion in the db.
Reason being, if I don't know that the words were 'relevant' to my content, then how do I know if I'm getting 'worthwhile traffic'? imho.
Example: You have an "A Quality" Site, but you donīt promote it. You wait that eventually visitors will come, because your site is really good.
A competitors site, with much less quality, spends time ( and money ) promoting it. Who is going to have more visitors?
I think that itīs much better to have good sites, that promote themselves, in order for visitors to reach them...
In a perfect world, you would just have to create a good site, donīt spend a dime on publicity, or spend time with link popularity and keyword density, instead of creating good content, but our world isnīt perfect, so thatīs not enough :(
Cyas,
Nuno Oliveira
The switch from GET to POST would increase the motivation to run automatic rank checkers. It would make databases like Overture's and Wordtracker's more valuable. It would increase the need for those of us with related sites to run our own niche search engines. It would increase the motivation to run short term Overture/Adwords style campaigns to find this kind of information. It would make me sad...
Calum
Knowing the amount of Web traffic a site has to process is vitally important to any organization (profit or non-). In order to manage the traffic and the infrastructure needs it dictates (for example, increased traffic necessitating more servers, or decreased traffic necessitating more marketing effort), the Web manager needs to know as much as possible about where the traffic is coming from and going.
You should be able to market a site without changing specific items on the website for just one audience( google for example). A good article is a good article, regardless of how the person who is going to read it, actually gets there.
Assuming of course they get there at all :)
Greektomi
>> Assuming of course they get there at all
Truly, that's the SEs job, not yours;)
With regards the switch from GET to POST... well I'd rather they didn't but if they do, as long as my competitors don't have kw's, then I'll cope.
There would still be a referral, but it would be from www.se.com/searchurl.cgi and not include the keywords. (the ? string wouldn't be present)
Bad move for the search engines. With the exception of Google, and Yahoogle, my own theme based site (which is free to all on theme businesses in my region) sends me more traffic than any of the other search engines ... including AOL & MSN. What would be the need for any search engine other than Google?
Why fewer companies...well, from what I've seen not all seo operations have the systems in place that could cope with this happening... .
What about sites/companies (particularly home hobby types) with little to spend - I really can't tell whether it would be a good or bad thing to happen
Search engines are welcome to use POST instead of GET, as long as they're willing to put money in my bank account [w3.org], send me a pizza [w3.org], or subscribe me to a service [w3.org].;)
Browsers don't [w3.org] like to bookmark [w3.org] POSTed results. People don't link to POSTed results. People don't email POSTed results.
GET is the Web friendly way to provide stateless search.
Of course this will probably count for nothing. The engines will copy google's text adverts approach and will want to bust the caches anyway.:(
Calum
