What’s the best way to proof a website with a client during the design + construction phase, online? I’ve noticed that sites get indexed quite quickly during proofing and begin to show up in google search results pretty quickly, often before the project is completed, which is not ideal.
I’d be interested to know what different menthols people use for proofing online.
The very simplest way to keep Google out is to add a cPanel password to your proofing folder. So if the link is http://yoursite.com/clientname, you would add a password to the clientname folder. Shouldn’t take but a moment, and then you have absolute certainty that the Google bot won’t get into that folder in the first place, and any links that follow it (if it did get in already) will require that password. You can also change the folder name every day (which will break any links that exist to its contents), although that gets to be a pain to maintain and communicate to your client.
Walter
On Sep 18, 2012, at 10:05 AM, Neil wrote:
Hi,
What’s the best way to proof a website with a client during the design + construction phase, online? I’ve noticed that sites get indexed quite quickly during proofing and begin to show up in google search results pretty quickly, often before the project is completed, which is not ideal.
I’d be interested to know what different menthols people use for proofing online.
Thanks for the fast reply. I didn’t realise that google bots couldn’t index inside that folder (duh, I know…). I currently use a folder like that on one of my sites for a secure download area, I never thought of putting a website in a similar folder for proofing purposes.
Thanks as ever.
Are you getting any closer to the ActionsForge Next pledge launch again?
No they cannot. But one way that they can find a link into that ghost folder is through something called “referrer leakage”. Whenever you click a link in a Web page, the server that receives the request also gets a reference to the referring page in the headers of the request. So your hidden file can become known to the world in that way. I’m not entirely clear how it gets from the referer header (yes, it is spelled that way in Apache for some reason) to Google’s ears, but I am assured by Those Who Know that it can and does. So longer answer, no, not if there are no links on the pages in your ghost folder, and if there are any, nobody clicks them. The old “tree in the forest” conundrum, I suppose.
Walter
On Sep 18, 2012, at 10:36 AM, Ernie Simpson wrote:
Can googlebots and the like find subfolders to which no published links
exist?