On 03/29/2017 09:04 AM, John Eells wrote:
> John Eells wrote:
>> R.S. wrote:
>>> Off-topic note: the document mention z/OS 2.2 can be ran on z9 machine,
>>> which is untrue.
>>
>> You are, of course, correct. The book has been updated but not the link
>> to the new level of it. Here is the new level of the book:
>> http://publibz.boulder.ibm.com/epubs/pdf/e0z3f113.pdf
>
> Talking with the person responsible for the book and the link, we
> found there is a bug in my browser. It's supposed to compare the page
> on the web to the cached one and refresh it if they differ when the
> tab is opened, but for some reason it did not. That is why I found
> the lower level of the book instead of the current one. Clearing the
> cache "fixed" it, and now the updated links are the ones I see on the
> page. I have never seen this failure before, so it was quite a surprise!
>
> Anyway, the book and the link to it are both current already.
>
That's not necessarily a "bug" in the browser. If the data served by a
website in response to a GET request doesn't explicitly give an
Expiration-Age and a Last-Modified time stamp for the data, a browser is
forced to use heuristics to determine a time after which a local cache
copy should be regarded as expired -- and different browsers may
determine different values for that expiration time in seconds.
If the browser thinks the expiration-age of cached data has not yet been
reached, it will not even attempt to re-validate whether the cached copy
agrees with the current web site, but just assume it is OK to reuse. I
have no knowledge of design choices made for specific browsers, but I'm
guessing that unless there is a bug, most browsers would avoid assuming
overly long cache expiration times unless there was something explicit
in the served data that suggested that was appropriate. When the
browser sees by the expiration time that it needs to re-validate the
cached data, there are ways it can request a new download iff the server
version has a later modification time stamp than the cached data time
stamp, but for this to work correctly does presume that both the browser
and server are dealing with time stamps in a consistent manner.
If you have reason to suspect the browser is showing an obsolete cached
copy for a web page, you can always force a re-validation and possible
new download by explicitly telling the browser to "reload" a web page.
If a web page is constructed from several "frame" pages, it might be
necessary to isolate a suspect frame into a separate tab by itself and
then force a reload on that one frame to get it to refresh.
Clearing cache will certainly get rid of any obsolete cached data, but
that may be overkill. In many cases just a simple browser "reload"
operation on the web page will be sufficient.
Joel C. Ewing
--
Joel C. Ewing, Bentonville, AR jcew...@acm.org
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
