More than 10,000 pieces of mail are dumped on the doorstep of the Copyright Office every day. Envelopes big and small, boxes tall and wide: They’re all opened, sorted, scanned, studied and stored in the Library of Congress’ Madison Building.

It can take months for one John Grisham thriller or Joyce Carol Oates novel to make it through the line. But copyright officials are hoping to cut the time and money spent on each item through a new Web-based system.

For now, however, the maximum processing time for most applications has increased from six months to eight. Thousands of copyright claims wait to be processed; last Thursday, employees were entering claims from March 14 into the system.

It’s partly a result of a difficult learning curve, and partly because the Web-focused system is still mostly processing snail mail applications, said David Christopher, the associate chief operating officer.

“We knew going into this that it was going to create a backlog. It’s going to take some time, and we knew that it would,” he said, adding that it will all speed up once more people use the Web to submit applications. “We are designed now to process electronic claims and the fact of the matter is we still are processing a lot of paper claims.”

With only 500 employees, the Copyright Office handles the registration for every literary, musical or visual work in the country. You don’t need to register to enjoy copyright protection, but that seal of approval can mean more money in a legal battle.

More than a million works get that approval every year based on their originality and uniqueness. Many are straightforward, but some are debated all the way up the copyright hierarchy. And about half come from “the average Joes out there,” Christopher said, people throughout America who want to ensure the everlasting uniqueness of an unpublished poem or photo.

A room in the Madison Building holds the records of it all on index cards in row after row of file cabinets. The registration of Superman, the ownership of Beatles songs: It’s all here. Hundreds of thousands of index cards fill the room, some written as early as 1870. In 1978, records went electronic, but otherwise not much has changed since that first card was filed — until now.

Since August, the office has employed a completely new approach, all centered on a unified computer system. Costing a little more than $50 million, the switchover took years to achieve. Now employees can scan paper applications, form a computerized record and affix a barcode to the work that tracks its every step. Less than a year ago, many of those same employees did most of their work without a computer.

The change has proved beneficial to those who submit copyright claims online: Registration now takes about a month for those who submit everything online.

But it takes up to four months when only part of the claim is filed through the “electronic Copyright Office.” And most of those who submit published works will still have to send a hard copy. However, sending in the application and payment electronically cuts the processing time in half.

A recent tour of the office revealed mounds of packages at every step of the way: off the truck, in big carts, barcoded and stacked. Copyright officials hope that in a few years, those mounds will shrink to a minimum. By September, they aim to have 15 percent of copyright claims at least partially submitted online; eventually, they expect that number to climb to 90 percent.

Until then, a system designed for the Web will have to handle claims filed by paper. At the same time, the new streamlined system calls for each employee to handle more tasks — and thus learn new skills.

For example, those who once only created the public record for a work now must also determine whether that work is copyrightable. They also are helping determine which published works should be included in the Library of Congress’ collection. Library officials still do some of that selection; every published work that enters the copyright office’s doors is considered.

The extra training has slowed things down, Christopher said. But eventually, the system will enable the Copyright Office to send “born digital” works directly to the Library and greatly speed up a cumbersome process.

Officials hope that eventuality is near.

“We’ve certainly gotten way past the learning curve here,” said Kim Brown, who manages the initial inputting, or “in-processing,” of materials. “Ideally we would like to ingest claims within a week of getting them. I think that’s possible.”