Let's try this again.

Although I have a hard time (as is obvious from the backlog, or lack thereof) keeping up with blogging, I figure I'll give it another go.

Part of this may be because my work e-mail and computer network refuses to accept I exist (again) so I am at somewhat of a loose end, unable to access pretty much anything.

I'm also going to re-design the main infomancy site and work on those articles.

(New Years Resolutions? Something like that, I suppose... I should also really master AJAX.)

An interesting idea for prospective employees

I was browsing around Facebook the other day when I saw hidden in their ads on the side the following text:

Let Me Be Your Librarian:
I'm a librarian looking for a library position. I have a MA and MLIS with strong programming and outreach experience. Please Click!

Clicking led me to the LinkedIn profile of one "MG Farrelly," a columnist/writer of things library-related, who apparently used to be a public librarian and is now looking to get back into a library for employment.

While I know lots of librarians browse Facebook (and other things) when they're at work, I'm not sure how many of them are likely to be in positions to hire. I think this is a fascinating application of the Facebook advertising system, and although I wish Mr. Farrelly the best of luck, I can't imagine his search will be too successful, given the current economic climate.

Still, it's a welcome change from all the ads I see which tell me I can get Google to pay me $44,444,440,550 an hour for doing nothing.

CDM 5 search problem update

Apparently the problem with CONTENTdm 5's searching was not just us.

I made a post to the listserv and learned that, in fact, several people were having the issue.

CDM support was in touch and provided the solution in the form of a new DMSystem.php file.

Apparently the problem was that "some erroneous null values were being passed to the search code and it wasn't handling them properly."

In any case, I copied the new file over and restarted the search service and it works fine now. Hopefully it will continue to do so.

Restarting Windows: It really does fix everything.

I restarted the server hosting our digital archives, and now CONTENTdm works just fine.

If it comes up again, I guess I'll have to figure out an actual real solution instead, but oh well. Works for me for now.

More CDM Woes

I thought my CONTENTdm tinkering-with days were over, but I was vastly mistaken!

Now I'm getting horrible PHP errors of death whenever anybody tries to do a search. Browsing still works fine.

My wonderful error is:

Warning: file_get_contents(http://127.0.0.1:56730/!/search?query=su%3Ahouses (za:f3 or 10.17.1.40)&sort=ti&display=$$DM_pa$$$$DM_ci$$&collection=/&suggest=1&facet=1&maxfacet=10&rsum=cr:da:fo:au:&facetinitial=cr:da:fo:au:&rankboost=&proximity=strict&priority=normal&unanchoredphrases=0&maxres=20&firstres=0&rform=/!/null.htm) [function.file-get-contents]: failed to open stream: A connection attempt failed because the connected party did not properly respond after a period of time, or established connection failed because connected host has failed to respond. in DMSystem.php on line 1753

Fatal error: Maximum execution time of 30 seconds exceeded in DMSystem.php on line 1753

What does this mean? For some reason the server isn't playing nice with searches. Why? I have NO idea.

But I have a meeting I have to go to now, so I will have to wait until later (when I should be working on my Tenure file. Whee) to find out.

Don't forget to index

So this one should be obvious, but I spent a while scratching my head before I remembered to do it.

After you've finished migrating (hurrah!) and all that, don't forget to index your collections! Indexing is required by CONTENTdm to let the search functions work, so if you haven't done it all your searches will come up empty.

At first I was afraid I was going to have to go through and re-do every single custom query on the site (due to some imagined change that made the ones from CDM4 obsolete). Fortunately, this is not the case. After indexing all the collections, the searches work fine.

Of course, if you have a ton of collections, it might take almost as long...

In any case, with this I'm now finished with the migration! I can move on to other things, like finishing my tenure file. Which I need to do very soon. After I've done that I'll probably go back and tinker with CDM5 some more, just for fun. I've always been interested in picking apart their "custom query" stuff and making it do horrible things for me.

CONTENTdm CQR pages showing up on Google

While trying to remember the URL to create custom queries in CONTENTdm, I was astounded to realise that CDM doesn't block access to the Custom Queries pages. This means that anybody can, once they know the URL, go to anybody else's page and build a page that links to their collections!

They even show up on Google, as a search for "CONTENTdm custom queries" will make obvious. I only saw 9 in the search results, which makes me wonder why Google is grabbing those and not others. However, once I knew the URL, any digital collections' site was fair game!

I'm torn on how I feel about this. It seems neat that I can now make web pages that will search the collections of libraries that I'm not even slightly affiliated with. I could theoretically build a mega-search tool that searches all the different CONTENTdm installations everywhere (though it'd probably be too much work to be worthwhile)!

On the other hand, it just seems like bad practice to let anybody into the back-end who isn't associated with your university, your library, your archives, what have you. For now I'll give CDM the benefit of the doubt and assume this is a purposeful thing (so that you can let professors or researchers access it, maybe?) and not an oversight.

But I still don't want it showing up on a Google search! Even if I am one of a very few who actually search for that.

The solution: robots.txt

A ton of people have created a ton of much more useful tutorials on the subject than I ever could, so I'll just link to one of those: http://www.robotstxt.org/robotstxt.html

In any case, I'll be adding this to the archives robots.txt file:

User-agent *
Disallow: /cdm4/cqr/

Of course, that doesn't mean people won't be able to access it, but at least it won't start showing up in Google...