Online research in philosophy

Some changes to the "new items" page:

1. This page will no longer show items with a publication date less than the current year, when there is a publication date. This was put in place because old material tended to clutter the page. I'm assuming that you don't care about an item being added to PP; rather, you want to know what's new in the sense of newly written or published. Some measures will have to be put in place to deal with N-dated items at beginning of year N+1. 
2. When the area filter is activated, items are no longer grouped by source and publication. This was creating too much clutter.
3. Also when the area filter is on, sorting is not done by date of addition to PP, but by date of addition to the area. This is so that one will not miss anything if one relies on the area-filtered version of this page to monitor new material. If items were sorted by date of addition to PP, an item classified n days after being added to PP could be missed if one does not scroll down n days upon visiting (items would make their first apparitions on this page at random points in it). Other constraints like not being published before current year still apply. 

I also changed the format and sorting of content alerts for material "in my areas". This is in fact a side-effect of changing the new items page, because the emails are just stripped-down renderings of this page.

I added a new heuristic to basic search mode: in keyword mode (i.e. when the search does not literally match an author's name), the relevance score of an entry is multiplied by two for each of the first two words of the query which matches the lastname of one of its authors. The result is that a query like this:

bourget intentionality

or like this:

david bourget intentionality

will almost always turn up what the person called "bourget" wrote on intentionality first.

Queries like this are the third most common kind of query, after pure author lookups ("david chalmers" type queries) and pure topic searches ("consciousness intentionality"). I'm hoping to improve their success without compromising the success of queries of other kinds.

For editors: added edited cats to "File under" menu.

BibTeX: changed ids to LastnameDateLetter format. also fixed encoding problems. everything should be TeX-encoded now.

Two major convenience changes:

1. Added "Category finder" to header. I was getting tired of going to a category to get access to this.

2. It is now possible to copy all the entries from a public category to a personal biblio, using an option on the personal biblios' option pages. Theoretically, the reverse is also possible, but this is currently disabled because the copy process does not support diffs for reversibility. If editors want to use this feature to populate their categories, they should let me know. I can arrange something.

A tiny but remarkable change: I made the search box focus on page load. This makes searching so much more pleasant.

Added a complete list of categories and editors here.

Misc changes:

• Focus of search box on home page only. This caused the page to scrollback to the top when changing sort order of listings, which was very annoying.
• Some old MindPapers items which have lost their links showed download counts calculated during the MP era. This is now fixed.
• Fix refresh of search filters. Often items which should have be added upon a filter change were not.

Two improvements in response to helpful reports by Daniel von Watcher:

• Added Zotero export option. This actually outputs EndNote data, but Zotero reads them fine.
• Fixed problem with BibTeX ids (removed newline before id).

I sort of neglected to post new changes here for a while. Been making multiple changes under the hood as part of the refactoring of PhilPapers for DIVRE. Visible changes I made recently:

• Fixed the bug which made browsers scroll to the top when exporting a single entry.
• Changed the duplicate filter not to count as different same-title-same-author works with different publication details (so likely reprint / alternate venues). One day we will have to restructure the database to keep track of editions / prints. Until then we conflate all editions of a work. This change explains a drop in the number of entries today because a number of duplicates were eliminated. 
• Getting ready to add a load of Library of Congress items.
• Fixed title bar of editor popup to show entry's authors and title.
• Fixed miscount of uncategorized item in weekly editor reports. 
• Fixed a lot more bugs (see bug report forum).

Today I upgraded the production server with the latest PhilPapers code. I made a bit of a mistake in the process, which resulted in something like 5 hours of downtime. It should be fine now, and there are a few notable changes:

• Areas of specialization have been added to profiles. 
• It is now possible to be logged simultaneously multiple times on PP (on different computers)
• A poll system has been added. This will come into use soon (more on that later).

I've fixed a number of bugs introduced by the last upgrades. Fixed problems with downloads (people getting pages of code instead of the files they wanted), reading lists, on-the-fly login when editing entries, and other behind-the-scenes issues.

Still patching up the holes introduced by the latest upgrade. Today I fixed problems with our cache which were making the app server a lot slower than it should be. We're now running about twice as fast as yesterday.

Fixed a bug which prevented some individuals with pro status from having their papers correspondingly upgraded. 

Fixed another issue with BibTeX exporting.  We now generate double { } titles so that PP's capitalization is preserved -- bibtex is really not good at capitalizing proper names etc.