Google Instant Pages sounds like it will be rendering the entire page, including images and other external resources. I wouldn't be surprised if it also executed JavaScript, fetched embedded iframes, and anything else that the page would normally do if you clicked on that link. I wonder if it would even follow redirections?

What is to stop a malevolent webmaster from performing redirects to nasty trojan or malware-infected pages if it detects the page is being pre-rendered? If that page contains flash objects, java apps, or other attack vectors on it, I'd think you'd be just as at risk as if you actually visited the site directly.

I could be wrong, though. I haven't thoroughly researched Google Instant Pages, but from what little I've found it seems it will be performing a full page render in anticipation of you clicking on the link to the page.

I did some Googling and apparently Chrome will send the following header when prefetching:

X-Purpose: instant

http://www.google.com/chrome/intl/en/webmasters-faq.html#instant

So it looks like it will be easy for me to block just as I have blocked Firefox prefetches.

Well, it is worth being called an idiot if it means I no longer have to keep track of the mail.google.com/hosted/* URLs and I can just use the generic Gmail login URL for everything.

They either added that functionality when they recently upgraded Gmail, or I somehow missed that functionality for all those years.

I just tested it and you are right. I swear it didn't used to work, though.

It looks like they are going to try to address that with the upcoming Visibility API:

http://code.google.com/chrome/whitepapers/pagevisibility.html

However, it seems to be JavaScript based which, at least to me, is not a desirable way to determine whether or not the page is being pre-loaded.

At least Firefox sent a "X-moz: prefetch" header which I used to ignore the traffic on those requests.

I added a simple check to my scripts long ago that detected Firefox prefetching and thew a HTTP 403 Forbidden status with a "Prefetching not permitted" message. It was straightforward to detect and block.

Hopefully Chrome either makes it easy to detect and block, or at least easy to detect.

The first issue is this is going to play havoc with traffic analytics and tracking. I'm sure Google Analytics will handle Chrome's Instant Pages just fine, but everybody else will have to figure out how to ignore Chrome pre-loads. I did some searching and they are adding a Visibility API to Chrome to allow authors of other traffic reporting packages to handle the difference. Hopefully the Visibility will be pretty straightforward and not require a lot of extra work.

The other issue is that this is going to eat up more hosting bandwidth. Popular websites that appear near the top are going to incur bandwidth usage that may never actually be actively used by the potential visitor.

Yes it does, at least for me. My Gmail-hosted work address is @mycompanyname.com

However, you have to log in to your @somethingnotgmail.com email account first, and THEN switch to your @gmail.com email address(es). So we're both right, in that you can do it, but Google really needs to improve the functionality to allow you to log in to @somethingnotgmail.com as one of your other email addresses when trying to switch. I also have noticed that when I log out of my @mycompanyname.com email address that Gmail now takes me to the generic @gmail.com login form, so I can not quickly log out and then back in which is lame.

Theoretically it should be an easy fix. All they would have to do is allow you to enter your full whatever@notsomethinggmail.com in the Username field and use behind-the-scenes logic to log you in to right account. For completeness and consistency, the Username field would also accept whatever@gmail.com and just ignore the "@gmail.com" portion of the username and log you in to your regular, generic Gmail address. Why it doesn't already work like that is beyond me.

I have both a Facebook account and G+ account, and more often than not I still use Facebook for the majority of my online interactions with friends and family. Most of my G+ circles are technical people (no family members, old school friends, or tech-illiterate friends yet).

Facebook has Friends Lists, which are pretty much the same as Google Circles. All Facebook has to do is make it more streamlined and easy to share posts only with certain Lists. The functionality is already there, but you have to click a few menus deep and then type in the names of the Lists (yes, you actually have to type out the List names ... heh). If they made a dropdown box off of the share post control that allows you to click on which Lists to share a post with then bam ... G+ loses one of its biggest advantages).

You do not need two separate browsers if you use the multi-signon feature in your Google accounts. That gotcha is you can't have both open at the same time, but you can quickly switch between them.

I have a Gmail-hosted work address and also a personal Gmail address.

When in Gmail, click on your email address on the top Google bar and select "Account Settings" and on that page under the "Personal Settings" section you will see a "Multiple sign-in" option. Set this option to "On". Do the same for all Gmail accounts that you want to switch between.

Then, when ever you are using Gmail you can click on your email address in the top Google bar and next to "Sign out" will be a link that says, "Switch Account". You use that to switch between your 2 or more Gmail accounts.

It isn't as cumbersome as my wall of text above would have you believe, and the only real drawback is that you can't view more than one inbox at the same time like you could in the old Gmail (I used to have each Gmail account in separate Firefox tabs before Google changed to the new mechanism).

It does not surprise me that Facebook would take such a stance, as they are a glorified data mining company.

I just hope the internet does not completely lose its wild west feeling. It is hard to not be concerned, though, as between walled gardens, paywalls, ever-more-draconian anti-piracy measures, bandwidth data caps, bandwidth throttling, multi-tiered internet, and cross-site tracking it is clear that corporations and government wants to change the web and internet as we know it. Then you have Google coming out and saying that they are banning accounts that do not use real names.

When IPv6 finally becomes mainstream and goes in to widespread use it will only get worse, in my opinion. With IPv6 every man, woman, child, dog, and toaster can have its own IP address. Not only would it be trivial to track most people by their IPv6 addresses, but as TV sets and other devices get connected it will be trivial to track and monitor the activity on these devices and tie that activity to their owners. The more tech savvy will be able to sidestep some of it if my prediction comes to pass, but the general public won't know enough or care enough to do it.

FOSS generally has bad UI compared to commercial applications. I use and love GIMP but it looks like a rented mule compared to Photoshop.

Historically FOSS looks more alien and complex, with a LOT of ability to customize the interface. I have always written this off as FOSS being created by bright individuals creating a tool to fill a need they have, but not really design it with other people in mind. Now FOSS seems to be trying to one-up the commercial apps in dumbing things down (see Gnome 3, Unity, and Firefox as examples). The saddest part is the loss of flexibility and configurability. Granted, this could just be FOSS shifting their focus to mimicking Apple instead of Microsoft.

I am not bashing FOSS, but for what ever reason commercial software just seems to generally have more polish and has a better time hitting the sweet spot between function and form. Look at PuTTY compared to SecureCRT as an example. Again, I love PuTTY, but SecureCRT is leaps and bounds ahead.

True, I am not saying that Facebook doesn't collect as much information on me as they can manage, but in scope they are amateurs compared to Google in that regard. It is not that I search for illegal or questionable content, but I just hate the idea of them keeping track of my searches. At least if I am signed out of Gmail/Google+ then their tracking is a bit more anonymous and doesn't associate my activity with my profile.

I think Facebook will end up benefiting from Google tying everything together under one umbrella. I can't speak for everyone, but I couldn't care less if my Facebook account got banned or closed. I primarily use it to socialize with some distant relatives and old classmates. It is not essential. If I lose my Facebook account it doesn't affect my email. Facebook doesn't track my Google searches.

Facebook may be a hated company, but at least they are relatively isolated.

It just doesn't make sense to put all your eggs in one basket. By using Google for everything you end up with a single point of failure.

Best-case scenario is that Google scares Facebook enough that Facebook improves their privacy controls and updates their friends lists to be more front-and-center like Google circles.

I currently have Version 6 Update 26 installed on all my home computers. Will that be the final release of the version 6 branch? I hate using x.0 releases until they work out any kinks or bugs, but I have also been bitten by having an old Java version installed (it is the only vector that has successfully installed malware on my computers).