- Others who have written about this problem have made the point that if your
application uses a permanent cookie to identify you, and if you view a page on
another site that "happens" to have links to your site, links you might not
even see, WebAccelerator will execute those links.
- They have
> -Original Message-
> From: Jochem van Dieten
> If users actually use Web Accellerator, Google has an even
> better indicator of page quality then the number of links to
> it: the number of hits and the time spend on a page. That is
> criucial information for a Trustrank. And if they
On 5/10/05, Barney Boisvert <[EMAIL PROTECTED]> wrote:
> ...No idea if that's how GWA works or not, but it
> stands to reason it might.
Speculation on the forum was running in that direction as well.
--
--mattRobertson--
Janitor, MSB Web Systems
mysecretbase.com
This is the way that Mozilla's prefetching works. No HTTPS, and
nothing with query strings. Of course, if you've SESed your URLs,
you're still hosed. No idea if that's how GWA works or not, but it
stands to reason it might.
cheers,
barneyb
On 5/10/05, Matt Robertson <[EMAIL PROTECTED]> wrote:
Incidentally some recent buzz -- supposedly from google staffers --
says that the fetcher will not pre-fetch any parameter with a '?' in
it. Sure would have been nice if they had mentioned this earlier.
Still not completely safe but I think that would googleproof pretty
much everything I do.
--
On 5/10/05, Marty Johll <[EMAIL PROTECTED]> wrote:
> Matt,
> Would you mind sending me a copy of the code you used?
Sure. See below. This is what Damien wrote. Nice and tidy.
Cheers,
--
--mattRobertson--
Janitor, MSB Web Systems
mysecretbase.com
Damien McKenna wrote:
> Just to explain the seriousness of this to people who don't have time to
> read the articles... GWA loads *every* single link on *every* page the
> user visits that isn't on an SSL site (i.e. https), including links
> after you have logged into a site, including links sayin
Matt,
Would you mind sending me a copy of the code you used?
Thanks,
Marty Johll
On 5/6/05, Matt Robertson <[EMAIL PROTECTED]> wrote:
>
> Thanks to everyone and Damien in particular for the heads-up and the
> code. I've planted it in a couple of /application.cfm's already.
>
> --
> --mattRober
r than the intended addressee(s).
-Original Message-
From: Jared Rypka-Hauer - CMG, LLC [mailto:[EMAIL PROTECTED]
Sent: Tuesday, 10 May 2005 20:35
To: CF-Talk
Subject: Re: Google Web Accelerator problems
I blogged about it the day this thread started...
http://www.web-relevant.com/blogs/c
I blogged about it the day this thread started...
http://www.web-relevant.com/blogs/cfobjective/index.cfm?mode=entry&entry=B506BE87-BDB9-5320-E5D41D3A198CB5AF
Let me know what you think of the post?
Laterz,
J
On 5/10/05, Paul Vernon <[EMAIL PROTECTED]> wrote:
>
> > Remember that it *does*, it
Paul Vernon wrote:
>
> I know this question probably belongs in community so I'd welcome any
> replies there but Is anyone else a little worried about some of the
> directions that Google seems to be heading off in these days?
"You have zero privacy anyway,"
- Scott McNealy, CEO SUN Micros
> I know this question probably belongs in community so I'd welcome any
> replies there but Is anyone else a little worried about some of the
> directions that Google seems to be heading off in these days?
Oh come now...what's wrong with a little world domination and the
obliteration of perso
> Remember that it *does*, it still prefetches pages after you have logged
> into e.g. a forum, which website spidering tools do not do.
If Google ever tie GWA into Googlebot by letting Googlebot have access to
Google caching proxies, that will have major copyright issues etc as all
that usually
> -Original Message-
> From: Thomas Chiverton [mailto:[EMAIL PROTECTED]
>
> 3) It's not doing anything the Google spider wouldn't if it
> could (see point 2).
Remember that it *does*, it still prefetches pages after you have logged
into e.g. a forum, which website spidering tools do not
lk
> Subject: Re: Google Web Accelerator problems
>
> Thanks to everyone and Damien in particular for the heads-up and the
> code. I've planted it in a couple of /application.cfm's already.
>
> --
> --mattRobertson--
&g
Dave Watts wrote:
>
> If it ends up really being that important, it will be too important for
> competitors such as Microsoft to ignore. Since Microsoft controls browser
> distribution, they are ideally placed to implement something like this in a
> way that best benefits them.
They already do: a
> I think they will continue to push it in one way or another
> because it is just to important to cancel.
>
> If users actually use Web Accellerator, Google has an even better
> indicator of page quality then the number of links to it: the
> number of hits and the time spend on a page. That is
On Sunday 08 May 2005 12:49, Paul Vernon wrote:
> just 2 or 3 for those people affected will make GWA very unpopular for
> people on capped services...
That wouldn't be very funny to find out would it...
--
Tom Chiverton
Advanced ColdFusion Programmer
~
Here in the UK, there are many boradband providers that cap download
capacity every month Increasing your download usage by a factor of even
just 2 or 3 for those people affected will make GWA very unpopular for
people on capped services...
Paul
~~
On Sunday 08 May 2005 07:33, Matt Robertson wrote:
> Me neither. What wories me is what happens when some ordinary user
I don't think non-techies will use it, because it isn't advertised to them
(yet).
> who thought it was cool and put it on logs into his intranet and...
I'm considering writin
Matt Robertson wrote:
> Me neither. What wories me is what happens when some ordinary user
> who thought it was cool and put it on logs into his intranet and...
> *whammo*.
Depends on how the intranet is organised. Most intranets I know
use RFC 1918 address space or SSL so the requests will not
Me neither. What wories me is what happens when some ordinary user
who thought it was cool and put it on logs into his intranet and...
*whammo*.
I'm not sure this is in fact a big deal to wory about. I can't see
Google keeping it as-is. Sure the RFC's say this and that but reality
just didn't s
On Saturday 07 May 2005 20:36, Jochem van Dieten wrote:
> As long as they also follow sponsored links :) I wouldn't install
> it myself though.
Neither would I.
Firstly, because I don't think people I visit will like a ten fold increase in
traffic.
Secondly, because it clicks things I don't.
Thir
Dave Watts wrote:
>> Don't blame Google for the errors of webdevelopers. The HTTP RFC
>> specifically says that GET request should not have lasting
>> side-effects:
>>
>>
>>In particular, the convention has been established that the GET and
>>HEAD methods SHOULD NOT have the significanc
On 5/7/05, Thomas Chiverton <[EMAIL PROTECTED]> wrote:
> Maybe this thing will honour robots.txt
Maybe someday but as of now it does not. Also regarding the comment
about the Google spidering capabilities, a spider can't get in behind
authentication. This thing can.
Yeah sure you're supposed to
On Saturday 07 May 2005 16:42, Dave Watts wrote:
> > 3) It's not doing anything the Google spider wouldn't if it
> > could (see point 2).
>
> However, we can limit spiders' access via robots.txt, or by requiring
> authentication.
Maybe this thing will honour robots.txt
--
Tom Chiverton
Advance
> 1) It's beta, get over it
Since it's a public and uncontrolled beta, that's irrelevant.
> 2) If the app is using GET for permanent changes, it's doing
> something it shouldn't (go read RFC2616 9.1.1)
Allowing clients to prefetch pages they won't use may be hazardous to your
site's performance
> Don't blame Google for the errors of webdevelopers. The HTTP RFC
> specifically says that GET request should not have lasting
> side-effects:
>
>
> In particular, the convention has been established that the
> GET and
> HEAD methods SHOULD NOT have the significance of taking an actio
On Friday 06 May 2005 19:34, Bryan Stevenson wrote:
> OK...that has got to be the stupidest piece of tech I've ever heard
1) It's beta, get over it
2) If the app is using GET for permanent changes, it's doing something it
shouldn't (go read RFC2616 9.1.1)
3) It's not doing anything the Google spi
Bryan Stevenson wrote:
> OK...that has got to be the stupidest piece of tech I've ever heard of...are
> they crazy??
Don't blame Google for the errors of webdevelopers. The HTTP RFC
specifically says that GET request should not have lasting
side-effects:
In particular, the convention has
On Friday 06 May 2005 19:19, Damien McKenna wrote:
> after you have logged into a site, including links saying "delete
> records", "send my boss a really nasty email", "order this book
It says it ignores Javascript confirmation and performs the action anyway ?
How does that work ? I have a JS func
Thanks to everyone and Damien in particular for the heads-up and the
code. I've planted it in a couple of /application.cfm's already.
--
--mattRobertson--
Janitor, MSB Web Systems
mysecretbase.com
~|
Logware (www.logware.us): a
> CGI["HTTP_X-MOZ"]
>
> '-' is illegal in a variable name.
>
> Hmm, how do you access such a variable name prior to CFMX?
You use array notation, as in your example.
> Are you sure that CF doesn't map '-' in header names to '_'?
Actually, it does, and I'd forgotten about that. You may also be
On 5/6/05, Dave Watts <[EMAIL PROTECTED]> wrote:
> CGI.HTTP_X-MOZ from within your CFML code.
CGI["HTTP_X-MOZ"]
'-' is illegal in a variable name.
Hmm, how do you access such a variable name prior to CFMX?
Are you sure that CF doesn't map '-' in header names to '_'?
--
Sean A Corfield -- http:
Thanks Rebecca.
--
Damien McKenna - Web Developer - [EMAIL PROTECTED]
The Limu Company - http://www.thelimucompany.com/ - 407-804-1014
#include
> -Original Message-
> From: Rebecca Wells [mailto:[EMAIL PROTECTED]
>
> No, you're both wrong. According to this
> http://www.mozilla.org/
> No, you're both wrong. According to this
> http://www.mozilla.org/projects/netlib/Link_Prefetching_FAQ.ht
> ml#As_a_server_admin_can_I_distinguish
>
> it should be:
> X-moz: prefetch
>
>
>
>
>
>
>
When an HTTP client sends an HTTP request header t
No, you're both wrong. According to this
http://www.mozilla.org/projects/netlib/Link_Prefetching_FAQ.html#As_a_server_admin_can_I_distinguish
it should be:
X-moz: prefetch
~
On 5/6/05, Damien McKenna <[EMAIL PROTECTED]> wrote:
> Here's the corrected CFML code. I had a slight typo X-)
>
>
>
>
>
>
>
>
>
Errr, there looks to still be a typo. I think
should be
Regards,
Dave.
~
Here's the corrected CFML code. I had a slight typo X-)
--
Damien McKenna - Web Developer - [EMAIL PROTECTED]
The Limu Company - http://www.thelimucompany.com/ - 407-804-1014
#include
~
> -Original Message-
> From: Dave Watts [mailto:[EMAIL PROTECTED]
>
> > > I would strongly recommend that you require SSL for any pages
> > > with the capability to destroy the fabric of society. I think
> > > I'd even recommend using client certificates in that case, too.
> >
> > How ma
> > I would strongly recommend that you require SSL for any pages
> > with the capability to destroy the fabric of society. I think
> > I'd even recommend using client certificates in that case, too.
>
> How many web forums are out there with lots of "delete user"
> links that aren't behind SSL?
> -Original Message-
> From: Dave Watts [mailto:[EMAIL PROTECTED]
>
> I would strongly recommend that you require SSL for any pages
> with the capability to destroy the fabric of society. I think
> I'd even recommend using client certificates in that case, too.
How many web forums are ou
> Just to explain the seriousness of this to people who don't
> have time to read the articles... GWA loads *every* single
> link on *every* page the user visits that isn't on an SSL
> site (i.e. https), including links after you have logged
> into a site, including links saying "delete records
OK...that has got to be the stupidest piece of tech I've ever heard of...are
they crazy?? I bet there's going to be one hell of a backlash ;-)
Thanks for the sample code Damien...I know I'll be adding that to
Application.cfm!!!
Bryan Stevenson B.Comm.
VP & Director of E-Commerce Development
El
Just to explain the seriousness of this to people who don't have time to
read the articles... GWA loads *every* single link on *every* page the
user visits that isn't on an SSL site (i.e. https), including links
after you have logged into a site, including links saying "delete
records", "send my bo
> -Original Message-
> From: Connie DeCinko [mailto:[EMAIL PROTECTED]
>
> [snip]
> http://www.mozilla.org/projects/netlib/Link_Prefetching_FAQ.html
You are missing something, as are many others. This has *nothing* to do
with Mozilla Foundation's support for page prefetching. This is Go
Suddenly the link to it was changed right before my eyes to the Google
toolbar. GWA link is dead too.
http://webaccelerator.google.com/
-Original Message-
From: Damien McKenna [mailto:[EMAIL PROTECTED]
Sent: Friday, May 06, 2005 10:53 AM
To: CF-Talk
Subject: OT: Google Web Accelerator
tp://www.mozilla.org/projects/netlib/Link_Prefetching_FAQ.html
-Original Message-
From: Damien McKenna [mailto:[EMAIL PROTECTED]
Sent: Friday, May 06, 2005 10:53 AM
To: CF-Talk
Subject: OT: Google Web Accelerator problems
FYI, it has been publicised today that there are problems with Google&#
FYI, it has been publicised today that there are problems with Google's
Web Accelerator and most websites that involve data processing links of
any sort. Read these for more:
http://37signals.com/svn/archives2/google_web_accelerator_hey_not_so_fas
t_an_alert_for_web_app_designers.php
http://davi
49 matches
Mail list logo
