IE6 Automatically Setting and "Last Checked" Meaning

  • Thread starter Thread starter AndyF
  • Start date Start date
A

AndyF

Hello,

I have two questions that relate to Internet Explorer
(version 6.0)

Both questions relate to cahcing and the specifics of how
it is handled in Internet Explorer.
We are implementing a Web based procurement system and
are tuning browser sessions to get the best balance of
network performance against download size.

1. The first relates to the Automatically setting of:
Check for Newer Versions of Stored Pages (Go to: Tools -
--> Internet Options. Click Settings

The help text states that "If you select this, Internet
Explorer will check for new content only when you return
to a page that was viewed in an earlier session of
Internet Explorer or on an earlier day. Over time, if
Internet Explorer determines that images on the page are
changing infrequently, it will check for newer images
even less frequently."

I would like to understand this algorithm better. It
appears that if some page content does not change over
time, then the browser will stop checking for updates
against the server. I would like to know how long it
takes to stop checking and what happens if the page is
updated at a later date.
We are concerned that changes to the site will not be
correctly relayed to users automatically.


2. When I look in the Temporary Internet Files directory,
there are "Last Checked" and "Last Accessed" dates on the
files.
Last Accessed is when the file was last used for display
from the cache.
What does "Last Checked" mean?
Is there less overhead in doing a check than there is in
downloading the file again? If so, how much less?
Or is a check basically going to the server and
downloading the file again?

Hope you can help with these.

Thank You,

Andy.
 
AndyF said:
Hello,

I have two questions that relate to Internet Explorer
(version 6.0)

Both questions relate to cahcing and the specifics of how
it is handled in Internet Explorer.
We are implementing a Web based procurement system and
are tuning browser sessions to get the best balance of
network performance against download size.

1. The first relates to the Automatically setting of:
Check for Newer Versions of Stored Pages (Go to: Tools -
--> Internet Options. Click Settings

The help text states that "If you select this, Internet
Explorer will check for new content only when you return
to a page that was viewed in an earlier session of
Internet Explorer or on an earlier day. Over time, if
Internet Explorer determines that images on the page are
changing infrequently, it will check for newer images
even less frequently."

I would like to understand this algorithm better. It
appears that if some page content does not change over
time, then the browser will stop checking for updates
against the server. I would like to know how long it
takes to stop checking and what happens if the page is
updated at a later date.
We are concerned that changes to the site will not be
correctly relayed to users automatically.


2. When I look in the Temporary Internet Files directory,
there are "Last Checked" and "Last Accessed" dates on the
files.
Last Accessed is when the file was last used for display
from the cache.
What does "Last Checked" mean?
Is there less overhead in doing a check than there is in
downloading the file again? If so, how much less?
Or is a check basically going to the server and
downloading the file again?
Click to expand...
1) I've never found those settings to work as stated. It seems it doesn't
matter what you choose, IE will behave as it wants.

IE has always been lousy at managing a cache. Even if you clear it using
conventional means, there are still hundreds of files cached in hidden
areas. It takes an app that's especially made to get rid of all the files
(and that goes for History files also). A little bit about that here:
http://www.enigmaticsoftware.com/cachesentry/index.html

I can't answer your question regarding last checked, etc. I empty temporary
files everyday.

It seems as though you are assuming that everything regarding IE works as it
claims to, and that's the exception, not the rule.
 
-----Original Message-----


1) I've never found those settings to work as stated. It seems it doesn't
matter what you choose, IE will behave as it wants.

IE has always been lousy at managing a cache. Even if you clear it using
conventional means, there are still hundreds of files cached in hidden
areas. It takes an app that's especially made to get rid of all the files
(and that goes for History files also). A little bit about that here:
http://www.enigmaticsoftware.com/cachesentry/index.html

I can't answer your question regarding last checked, etc. I empty temporary
files everyday.

It seems as though you are assuming that everything regarding IE works as it
claims to, and that's the exception, not the rule.


.
Click to expand...
Thanks Smoker,

I am working on a 'bought-in' product that seems to
further complicate things.
If I sit and watch the cache as I go through the system,
it appears to be using cached files, but our network guys
are sure that every page brings back every item from the
server.

I got some help from Microsoft about file expiry dates on
the server, but then they wanted to make me pay for
support, so I gave up.

Andy.
 
AndyF said:
Hello,

I have two questions that relate to Internet Explorer
(version 6.0)

Both questions relate to cahcing and the specifics of how
it is handled in Internet Explorer.
We are implementing a Web based procurement system and
are tuning browser sessions to get the best balance of
network performance against download size.

1. The first relates to the Automatically setting of:
Check for Newer Versions of Stored Pages (Go to: Tools -
--> Internet Options. Click Settings

The help text states that "If you select this, Internet
Explorer will check for new content only when you return
to a page that was viewed in an earlier session of
Internet Explorer or on an earlier day. Over time, if
Internet Explorer determines that images on the page are
changing infrequently, it will check for newer images
even less frequently."

I would like to understand this algorithm better. It
appears that if some page content does not change over
time, then the browser will stop checking for updates
against the server. I would like to know how long it
takes to stop checking and what happens if the page is
updated at a later date.
Click to expand...

I never had the patience to try to figure it out.
I stopped using Automatically long ago when they fixed
the main performance problems with Every visit...

We are concerned that changes to the site will not be
correctly relayed to users automatically.
Click to expand...

When the content seems stale or inconsistent the user is expected
to press F5 (click Refresh) which is equivalent to a one-time
application of Every visit...

Ctrl-F5 is analogous to a one-time clearing of the cache for the
content of this particular page. Not really equivalent I suppose
since old content which was cached would still be old cached content
if it wasn't referenced by the new version of the page.

2. When I look in the Temporary Internet Files directory,
there are "Last Checked" and "Last Accessed" dates on the
files.
Last Accessed is when the file was last used for display
from the cache.
What does "Last Checked" mean?
Click to expand...

It would be the statistic you would be interested in to understand
Automatically. If Last Checked == Last Accessed then the IE
cached version has been checked with *a* server. If Last Checked
is less than Last Accessed then IE would have pulled its copy
from the TIF without checking with a server. Compare what changes
if you load the page while in a Working Offline state (Alt-F,W)
and what changes if you load the page while online with Every Visit...
I think it may also be useful to keep track of the actual local file's
Creation date (which unfortunately is not shown by either the TIF viewer
or the TIF file properties display.) I don't trust the Last Modified statistic
for this purpose because I see None even for really old accesses.

Notice my emphasis of "a server" above. A problem with checking
can occur if there is an intermediate cache. IE gets the Ok from the
cache and uses its cached copy but that only guarantees that the user is
going to be shown the current content if the intermediate cache is also
checking reliably with the host server.

BTW a very useful tool for capturing a snapshot of the TIF and
comparing it with another one is David Pochron's CacheMon2.

Is there less overhead in doing a check than there is in
downloading the file again? If so, how much less?
Or is a check basically going to the server and
downloading the file again?
Click to expand...

Depends on what you are calling overhead.
Byte-wise there is definitely less overhead.
Timewise for small files it's probably not enough different
given the risk of having the wrong content.

IMO if overhead was a concern for content providers I think
that they should not be marking so much of the big stuff
non-cacheable. That trend, which has really ruined the possibility
of offline browsing using the TIF, seems to be based more
on copyright and advertising revenue than any other concerns.

Hope you can help with these.

Thank You,

Andy.
Click to expand...


HTH

Robert Aldwinckle
---
 
Back
Top