Big Domino View Search Test Results
Remember the Big Domino View Search Test? Well, the results are in. I can feel the anticipation in the air!
Recap: Domino views on the web have a meta tag, called "robots", with the value noindex. This is why search engines don't index Domino views (not to be confused with Notes view indexes, which are a complete other kettle of fish). Sometimes this isn't what we want. This test was to see if we could over-ride the noindex setting with a duplicate robots meta tag that we add ourselves.
I created a simple DB to test this and have been waiting for Google to index it. I knew it had because the sole document in there contained the line "Big peacocks sailing on submarines don't make dinner for nobody" (analyse that!), which, as I hoped, appears nowhere else on the web and is the sole result in Google.
So, the database has definitely been trawled by Google. Did it index the two views - one with only the noindex tag and one with a conflicting tag. So that I could test this I'd added each view's ID to the page - again hoping these strings would by unique.
As expected the NoIndex View can't be reached by searching for 766c7254682ecf8b80256feb00300e69. Can we find the Index View by searching for b9af01f92bea128e80256feb00305d10 though? No.
Wondering what I am on about? This is just a long-winded way of me saying that you can't over-ride Domino's telling search engines to not to index pages created using a $$ViewTemplate form. Not the end of the world, but it's another example of Domino giving us little control over "our" HTML.
Skip to the 
You can subscribe to an individual
Hi Jake,
The simple conclusion on that:
have just $$ViewTemplateDefault with the following formula in the HTMLHead (and an text in the body: "if not.. redireceted click here...":
dbURL := @WebDbName;
"<meta http-equiv=\"refresh\" content=\"0;URL=" + dbURL + "?OpenDatabase\">"+@NewLine+
"<meta name=\"MSSmartTagsPreventParsing\" content=\"TRUE\">"
I've stopped using other $$ViewTemplates. Instead I have a nice configuration document, where I specify the view to embed, a menu to appear, a css to use, extra text ect. One (or more) configuration per view. So I open the document instead of the view and have full control.
:-) stw
want the template? PM me.
:-)stw
Jake,
Don't confuse a visit by the Google bot and non-appearance in the search results to be the be all and end all. The google bot crawls pages and stores the results, but the indexes are only refreshed periodically (depending on load and how often Google thinks your site changes). When I've looked in the past, this can take from a few minutes (in the case of heavy traffic sites) to several weeks for the results to appear in the indexes.
Also, check your stats to see if the google bot accessed the view AND the document. Again, the google bot often only reads what it knows exists, and comes back later to follow links from that content. I think they do this as a), they know that most content doesn't change very frequently, and b) they can schedule new content crawls when they know other workload is low.
Finally, if you've got '?' in your URL's for the views (and linked documents), Google probably won't index the view anyway. You need to use '!' so that Google doesn't think your causing a dynamic query for the page.
Dave. There are several large assumptions made in what I've "concluded". However, I think it's safe to say that Google has indexed the database. Well, at least the sole document is in the index. That's for sure. Maybe Google indexed this and not the views. Would seem strange though. Although nothing surprises me with Google.
The thing about "?" in URLs isn't true any more (was it ever?). Take the document from this test database. The link to the Google search in this entry ("Big peacocks...") returns a URL to it that ends in ?OpenDocument. It's still true of some search engines, but not Google.
I think that in this case you'll find that it's Google that's done you wrong, so to speak. They seem to have fixed the intelligence that previously prevented them from indexing Domino sites properly, but they've done it by essentially ignoring views. The view is used as a "link source" to generate the document index, but note that they've "thrown out" the view in the links. Regardless of the view the document link is derived from, they will use "/0/" (which has broken any number of Domino apps that relied on form formulas to render "safe" HTML to users). My take on this is that Google have taken advantage of someone with Domino expertise to handle the nsf-in-the-url case, and that someone has made a lot of assumptions about what Domino applications are supposed to look like. You know -- views on the web should look like views in the Notes client, and so forth. Obviously not anyone who's spent any time skulking about on this site....
Good point Stan. I'd not noticed the view "name" being replaced with "0". Odd. Troubling as well, if you rely on form formulas, as you say.
Hmm . . . checking {Link} I see that the 'noindex' directive means 'don't index this page, but follow the links'. If the bot doesn't index a view, does it really matter? What matters, surely, is whether the bot indexes the destinations that a view links to - and this shouldn't be affected by use of the 'noindex' directive. It's the 'nofollow' directive that we should be concerned with, IMHO, and this isn't generated by Domino.
Simon. More often than not this isn't a problem. I've never needed a view page to be indexed. It wasn't until somebody who did asked me about it that I wondered whether it was possible.