Am I just overlooking the link to the recording?

Click "Show menu" - there is the recording link and the attachment.

Got it, thanks. Learn something new everyday.

Despite my best intentions, getting up at 3:00 AM in the morning in Australia to watch the presentation was just too hard this morning.

As per the Mike Wilson experience, and even after reading your instructions, the SlideSix link to the presentation still takes a bit of poking around to find (not exactly the most intuative interface). Now I find I need some login other than my Adobe username and password to see it. Any ideas? Why is a username and password required to see it anyway?

What is this presentation coded in? it's not exactly selling itself. Compare to the likes of You Tube/Vimeo etc even Adopbe Flash. Is it 'new' Flash? All I got was a bunch of old-school looking slides - no sound, no video and lots of waiting...waiting? What could I be doing wrong?

Bill - the Flash app above is _just_ the slide deck. The recording is available when you click Show Menu and then click "View recorded ..." icon. As I said to Mike above - that's what you click for the recording.

Unfortunately - as Chris has shown - the recording is password protected. I'm working on that right now. Being the weekend it may be a while though.

Thanks so much, Ray! Do you think it would be possible to als have your demo files for download? It would complete the experience.

Click on Show Menu and click the paper clip icon I think it is - it will let you download the files. Everything but the Zeus preview. ;)

Wow, that was fast! Thanks. I guess, it is not as intuitive as one might have hoped at SlideSix. Or it's my intuition that fooled me. Again. Have a fine weekend!

Guys, the recording URL should be fixed. You do need an Adobe.com login (which is free), but it should work.

Cool. All good now :)

Hi Ray,

I have a love /hate relationship with CF9 solr
I have cf9.01 with updater 2 installed. I'm not doing anything extraordinary:
1. Create a collection based on a db query
2. Index it
3. Search it

Some how it doesn't work as expected:
First the summary field is capped at 100 bytes even if the search includes: ContextPassages="3"
ContextBytes="2000",

2. I would expect the result context to include the first paragraph where the search term was found. Currently is using the summary and if the search term is not in there the context is empty

Am I missing something obvious? I have looked into the solrconfig.xml file but couldn't find anything obvious

Tanks
Mike

I have some more information. Not sure if it helps or not:
the 100 characters for summary is hard coded only for queries.
For files is 300 (as per Adobe help), also hard coded, meaning that ContextBytes has no effect.

For both cases the context is based on the summary which gets the 100(300) bytes from the beginning of the document. If the search item is not found in the first 100(300) characters the context is empty which pretty much makes it useless.

Imagine if Google searches will give just the beginning of the page with no context information about what you are searching...

Thanks
Mike

I do not believe you are right at all. You can be right on summary, but context can be found anywhere in the document. I did a simple text with a txt document for a book. It found a test pattern from various parts of the book and the context clearly showed it.

Also, you want to ensure you enable term highlighting. This is done in the CF Admin on a per collection basis.

Thank you...

That was it. By enabling it in CFAdmin the context now shows highlight terms properly.

Is this specified anywhere in the docs. I was doing everything in the code and didn't know about this.

Can it be done programmatic-ally?

In any case thanks again and if Adobe did anything right in the few months was getting you back as Evangelist.

Best Regards
Mike

I think - stress think - it is the one thing you can do programatically. Maybe via the Admin API, but not via cfcollection.

Is this presentation supposed to still be available? Attempts to view the pre-recorded presentation throw an error on Adobe.com

Adobe.com had issues yesterday. I'd try again. (Although I see slidesix is down now too - pinging him.)

Any way of filtering (other than file extension) files in a directory to be added to a collection? I have the filenames in a database field and like to make a collection based on this field using a query.

You can use cfindex to add one file at a time.

Hi - my company is switching from a hosting plan that has CF 8 (Verity) to CF 10 (Solr). I set up my search collections on the CF 10 server the same way I did the CF 8 collections but noticed something odd when I searched against them. The context from the Solr search is missing or very limited even on results that have a summary and are high in the result set while the Verity context is always there. I've tried destroying and rebuilding the collection but that didn't work. I thought it might be similar to Mike's issue (09-19-2011 at 09:39 AM)but that doesn’t seem to be the case either (or at least I don’t think so) and I’m pretty sure that the collection is set up for highlighting based on what I see in the search results. At any rate – when I go into the CF Administrator to try to get it to enable highlighting – I keep getting this error “There was an error enabling highlighting for this collection. An error occurred when performing a file operation write on file C:ColdFusion10cfusioncollectionscontent/conf/solrconfig.xml.” (the error appears on screen lacking slashes where I would expect them) When I look at the Solr config file – it seems as though highlighting is enabled – at least according to what I see on http://wiki.apache.org/solr... .

I’m not sure if it has to do with how the collection was created, how it is indexed, or how it is being searched – it’s just a mystery to me at this point.

Hmpth. You got me there. It sounds like something is locking the solr file and preventing CF from using it. Normally I'd say delete/recreate the collection, but you tried that. It may be best to reach out to tech support.

Thanks Ray - I'll try that.

BTW - Your blog is great and I've learned a lot from it.

Thanks Scott. :)

I started looking at this again this morning and wanted to explore something my coworker found last week - an Adobe link that suggested updating the XML files - http://help.adobe.com/en_US.... When I logged into my server, I noticed that my co-worker who was helping me look into the issue had done the changes to the C:\ColdFusion10\cfusion\jetty\multicore\template\conf directory so I tried my search again and it was still missing a lot of context. I was searching against my “content” collection that is stored in C:\ColdFusion10\cfusion\collections\content\conf and it still had the old XML content. I stopped the service, modified the XML documents, restarted the service, purged the collection, then re-indexed it using some code I wrote (still not sure how to re-index a collection that is based off cfquery through the CF Administrator), then retried the search. It works now. :)