Welcome to WebmasterWorld Guest from 22.214.171.124
If there's no cache available, why did it give me the option to view it in the 1st place?
Google is really starting to look broken from where I sit.
I see that on some IPs and not others. I tried the same search on 6 different IPs, then clicked to view the cache of one particular site. On two IPs there were no results available. I guess that not every machine has the full index all of the time.
I also clicked the cache for www.domain.it/folder/page.html and was given the cache for www.keyword.domain.it/page.html and the cache was announced as being that of www.keyword.domain.it/page.html toom, where both URLs are two ways to access the same content on the same physical server. Google knows they are duplicates and has only one cache for the pair.
[edited by: g1smd at 2:04 pm (utc) on Feb. 24, 2005]
Google has cached the new version of the page, sans email address, many times since then. The Google cache does not show the email address as being on the page any more.
However, if you do a Google search for the email address, then you'll get one result: the page that it used to be on, but get this... the email address, the one that is not on the real page, and not in the cache is printed in the snippet.
How's that work?
In case you think that is an isolated example, let me tell you that I know of FIVE sites where this has happened.
I doubt it would require all of this to fix the 1969 issue. I don't think the 1969 issue was all that big (it just means they couldn't determine the date).
Although you have to wonder - Why couldn't they determine the date. I know that this digs up an old topic - but what would happen if their database ran out of space? would new data overwrite the old (hence the date for a page going MIA).
I'm not sure how many of the IP's this is happening on - I don't really care to be honest, but ever search I've done over the past 2 days has yielded these errors, and i know I've been hitting more than 1 IP, so it doesn't seem isolated.
Why doesn't google just take off that cache link temporarily until they resolve the issue. It looks tacky and will probably only perpetuate rumors of the overloaded database theory.
How can you update a cache that doesn't exist?
Problem solving 101: Before fixing something, be sure of what cuased the error in the 1st place, then set rules in place to ensure the same error cannot happen again.
With that in mind - What would cause Google to have lost the string or piece of information that kepts record of the date, on a wide scale?
Another question to ask is: Just how important is the date anyway?
I think it's quite important. If I am reading some material, I like to know when it was published - if that isn't available on the page, then at least the date it was last crawled is available (well sometimes, but not now)... it's not the published date, but if it's a news article it will at least let me know if it's a week or so old or whatever.
I just checked again, and in the last few minutes the Feb 23 @ 03:00 UTC cache date has updated to Feb 24 @ 03:00 UTC (or sometime about that hour). This happens every night sometime betweeen 22:00 UTC and 01:00 UTC.
I only see missing cache in a few sporadic IPs, not widespread.
If I hit reload on a cache dated 23rd Feb it changes to 17 Feb.
Every time I hit reload again, it changes to the other one.
This is for a site that I expected the cache date to update to 24 Feb, sometime about now; and still expect it to do so in the next hour or two.
I wonder if the cache is reverting, simply because the new page uploaded 24 hours ago (version 3) has LESS on it?
<edit>For my other "cached daily" website there is no cache on this particular 64... datacentre; even though the "Fresh date" just updated from 22 Feb to 23 Feb just an hour or so ago. Looks like this IP has a very old, incomplete, set of cached pages.</edit>
[edited by: g1smd at 12:56 am (utc) on Feb. 25, 2005]