Yesterday I was searching the New Yorker DVDs and found an article by Jay McInerney. I enjoyed Bright Lights, Big City, so I went back to the search to see what else he'd written for the magazine (check out Philomena in the Dec 25, 1995 / Jan 1, 1996 issue). But I couldn't find his name in the Authors list. It turns out this is because when his name was entered into the database, he was entered as Mclnerney, Jay rather than as McInerney, Jay. The font in the viewer (as well as the New Yorker's titling font) makes an 'I' (upper-case 'eye') look just like an 'l' (lower case 'ell'), so it's hard to see the error unless you look in the database itself. (In fact, it may be hard to see in this post depending on what sans-serif font you're reading this in!)
The fix is a very simple one if you have the sqlite3 interface to the database:
> sqlite3.exe ny-sqlite-3.db sqlite3> update Authors set Author="McInerney, Jay" where AuthorID=13374; sqlite3> .quit
It's probably a good idea to make a copy of the ny-sqlite-3.db file before manipulating the database. On a Windows machine, this file is in C:\Program Files\New Yorker Viewer\.
I did send email to the New Yorker DVD technical support, but I don't know if this is the sort of thing they'll fix on their own or not. I hope so. Luckily, we can fix it ourselves!
A few hours ago I wrote about a script I wrote that displays your New Yorker DVD reading lists in a convenient format. I just finished writing a similar script (again using Python and the pysqlite python module) which dumps the SQL statements you'd need to restore a reading list to the database. To restore the data to the database you'll also need the command line sqlite3 tool.
Use the script, reading_list_dump.py, to periodically dump your readings lists to a file:
$ ./reading_list_dump.py CSS > css_reading_list.sql
The file you created (css_reading_list.sql in this example) will contain SQL like:
INSERT INTO ReadingListEntry VALUES ( 700, 7, 181437 ); INSERT INTO ReadingListEntry VALUES ( 701, 7, 154899 ); INSERT INTO ReadingLists VALUES ( 7, 'CSS' );
To insert them into the database, close the Viewer program and use the sqlite3 command line program to upload the file:
$ sqlite3 -init css_reading_list.sql ny-sqlite-3.db
One important note: The table that's being updated (ReadingListEntry) has a primary key field that has to be unique in the table. It's the first number in the INSERT statements above (700, 701, etc.). Before running the sqlite3 command, you should make sure that there are no rows in the ReadingListEntry table that match the data you're inserting. If there are, you'll need to edit the dump file and change those numbers until they don't conflict with the existing table.
To find out what the largest value in the ReadingListEntry table is, run the following commands:
$ sqlite3 ny-sqlite-3.db sqlite3> SELECT ReadingListEntryID FROM ReadingListEntry ORDER BY ReadingListEntryID DESC LIMIT 1; 585
All the values for the first field in the dump file for the ReadingListEntry table should be larger than this number.
The New Yorker DVD's allow you to create your own personal reading lists, which are lists of articles that make it easy to access the articles you want to come back to. You can even restrict your searching to your own reading list. The DVD comes with several reading lists already loaded, including one for all the articles mentioned in the hardcover book that comes with the disks.
As it turns out, the database used by the viewer program is a SQLite3 database. It's a public domain database, and the New Yorker hasn't encrypted it so you can view the data in the database with your own custom program, or with a SQLite command line tool available from the SQLite3 web site. You can also extract the data for your reading list, in case something goes wrong with the database and you need to reinstall it.
I wrote a simple Python script, reading_lists.py that allows you to generate a formatted display of all the articles in your reading list. To use it you'll need a standard Python installation, and the pysqlite Python module.
To use is, you'll need to find the database file, which is at C:\Program Files\New Yorker Viewer\ny-sqlite-3.db in a default Windows install. I'm not sure where it installs on a Mac, but the filename is probably the same. Then just run the program specifying the database file, the reading list, and the other options the program offers. The result is a list of the articles complete with abstract and keywords.
Here's the usage statement:
Usage: ./reading_lists.py [-h] [-v] [-c columns] [-d database ] [-r list] [-y year] -v Verbose -h This help page -c columns Number of columns in the output (72 defautl) -d database Full path to the database file -r list Name of reading list (escape spaces or quote list) -y year Year cutoff (negative means article before listed year)
and a sample of the output:
------------------------------------------------------------------------ The Financial Page: The Catastrophe Problem. Surowiecki, James The Talk of the Town Jan. 10, 2005 (Disk_1_1998_2005)
THE FINANCIAL PAGE about the difficulties presented by natural disasters and catastrophes for the insurance industry. . . Writer notes that, in dollar terms, the recent Indian Ocean tsunamis impact was relatively small. . .the affected countries are relatively poor and the businesses there are less likely to have insurance. . . This past year was the most expensive in history for insurers. . . Discusses the actuarial nature of the insurance business: its about distributing risk over a large group of people whose behavior is collectively predictable. . . Catastrophes are not nearly as predictable. The same goes for man-made catastrophes such as September 11th. Insurers can make forecasts as long as they take a much longer view. Notes that even after September 11th, the number of companies that purchased terrorism insurance was relatively small. . .
Keywords: Actuarial Businesses; Hurricanes; September 11th, 2001 (9/11); Catastrophes; Insurance; Terrorism; Asia; Natural Disasters; Tsunamis; Business;
Note that if your database did get corrupted, you'd still have to manually recreate your reading list. I'll probably write a second Python script that will generate all the SQL INSERT statements necessary to recreate your reading lists if something went wrong with the database.
For Christmas this year I bought myself The Complete New Yorker, which is an 8 DVD collection of all 4,109 issues published between February 1925 and January 2005. It's got a nice interface, is reasonably searchable, and the scans are good enough to read them on screen without strain. Printouts of the pages don't look as nice as the original magazine (or quite as nice as on screen, oddly enough), but it's perfectly acceptable.
Because I run Linux and the proprietary software only runs on Windows and Mac OS/X, I run the software through VMware, which allows me to run a virtual computer with Windows installed.
I spent a few evenings trying to get the DVD's backed up to a hard drive so I wouldn't have to continually swap the physical media to access all the content. I was finally successful using Alcohol 102% to back up the data. Then, using Daemon Tools, I can mount the DVD images as though they were physical media and the New Yorker viewer works.
There's still something not quite right about this, though. Even though Daemon Tools allows you to have as many "drives" connected as you want, the New Yorker viewer will only recognize one of them. That means you can't just connect them all and have the viewer grab whatever content you're looking for. I'm not sure if this was intentional or not, but it's unfortunate. Instead, when you choose content that's on one of the other disks, you have to click the Daemon Tools icon and choose the proper image.
The other quirk is that the order of operations appears to matter. To make it all work, you need to mount one of the DVD images with Daemon Tools, start up the New Yorker viewer, and then choose content that's on a different disk than the one that's mounted. If you try to access content on the currently mounted disk, the reader never opens, and if you try to mount a DVD after you've started the program, it doesn't work either.
One other note -- VMware has the ability to connect disk images to what Windows interprets as physical drives, but this doesn't work with the New Yorker DVD's. It's not completely clear why, since Daemon Tools is basically doing the same thing from within Windows, but I think it's either that VMware isn't completely emulating a DVD drive when using an image (the drive looks more like a CD-ROM drive in Windows), or that the copy protection on the original DVD's isn't emulated.