PDF Collaboration: WebDAV In Action – The HTML Communications Standard

Editor’s Note: This article is part of the PDF Collaboration: WebDAV In Action article

we can share PDF files on the Web, we must first find a way to get them
there. Despite promises in numerous marketing brochures, putting data
online is not often an easy matter. Absent any support from CGI scripts,
the online data must usually be uploaded to the Web server using an FTP
application; the normal browser will not suffice.

With Acrobat, there are two possible solutions for
uploading PDF files to a Web server. One is WebDAV, which is an extension
of the HTTP 1.1 protocol. It is an interpretation of the IETF working
group that developed a proposed standard entitled ‘RFC 2518: HTTP Extensions
for Distributed Authoring.’ Adobe, among others, has adopted this approach.

WebDAV (Web Distributed Authoring and Versioning) extends HTTP to add
the capability of securing data on the server. Further, members of a team
can use WebDAV to work on the same document at the same time, without
being in the same place. Only an Internet connection is required.
The shared access is implemented by functions such
as file locking and version control. The locking feature allows a user
to temporarily block access to a file while he or she is working with
it. Once the changes are completed, it is unlocked again. Locking and
unlocking happen automatically, controlled by WebDAV, to avoid a ‘collision.’

It is not necessary to maintain a network connection during the time the
lock (called a ‘persistent lock’) is applied to a file. Thus, a file can
be opened online and edited offline. Subsequently, the changes are ‘written’
to the server.

WebDAV also provides for the association of properties
with documents. These properties are metadata encoded as XML. WebDAV distinguishes
between ‘dead’ and ‘live’ properties. Live properties are generated by
the server itself, including such things as creation date and date of
modification. Dead properties are name-value combinations that incorporate
a URL and XML coding. In the case of Acrobat, these are online annotations.

But WebDAV does its work just as well in an intranet,
and it can be implemented on a network server. Since WebDAV is dealing
with ‘pure’ HTTP, up-to-date firewalls and proxy applications do not pose
much of a hindrance.

Many software products have recently been developed
for WebDAV or have added support for this new standard. For example, Cadaver
is a command-line client for DAV, which can be used for uploads, downloads,
copies and moves between client and server. PyDAV is a DAV server implemented
in Python, and a Python DAV client library is being developed. The Jakarta
Slide Project was done with WebDAV. This collection of modules is leading
toward a content-management system based on WebDAV. Another open-source
tool is Sitecopy, which uses WebDAV and FTP to simplify Web site management.

For the Mac O/S, the DAV ? Goliath project is in the running. There
are four separate projects to develop servers and server-management solutions
based on WebDAV. On the server side, WebDAV is supported by market leader
Apache (via mod_dav) and by Microsoft Internet Information Server 5, among
many others.

A long list of commercial products are using the standard
as well: Adobe GoLive, Apple Mac OS X, Xythos (a WebDAV server), Macromedia
Dreamweaver, Excosoft, Xerox Docushare 2.0 and Microsoft Office 2000,
to name just a few. The implementation in the universally-accepted standard
HTTP appears very promising — if the future inclusion of WebDAV on the majority
of Web servers could be assured and if several problems get solved. Although
WebDAV is still in its development stage, it is clear that the market
need is great. Even market leader Microsoft, in spite of its own technologies
such as FrontPage, supports WebDAV. If it becomes possible for ordinary
users to securely maintain data and documents on network servers and Web
servers, that will be good news indeed.

You May Also Like

About the Author: Bernd Zipper

Leave a Reply