Oxford English Dictionary software comparison

Skip to first unread message

Robert Holmgren

Mar 2, 2003, 11:46:54 PM3/2/03
Recently I had an opportunity to observe the new OED Version 3
software in use. I own Version 1 of the OED Second Edition software,
and my impression is that the new Version 3 is not much improved --
that in several respects it is, in fact, inferior. Perhaps
more experienced users would care to respond to the
following observations (admittedly based on a very brief
experience)? I'm trying to discover whether there is any
good reason to upgrade from v1 to v3 (at a cost of USD

OUP seems to have transformed the engine into bloatware on a
massive scale. Whereas v1 fit on a diskette (just under 2
million bytes), v3 is now about 1 billion 70 million bytes,
not including dictionary data (about 530 times larger!).
This amounts to the biggest application I have ever seen --
much bigger than any modern operating system -- and *NOT*
including the dictionary data!

Ordinary lookups seem slower, and to consume much more
random access memory. Speed declines markedly when other
applications are running. In contrast, v1 is lean and fast.

v1.10 (but not v1.13 or v1.14) could be installed entirely
on a hard disk. No "validation" of the data disk was
required. v1.10 could be networked (OUP says that v3 cannot
be networked). v1 ran on a Macintosh.

Installation and operation of the C-Dilla security system in
v3 is not documented by OUP. I have read genuinely scary
stories about C-Dilla, including that it writes to areas of
the hard disk MBR (Master Boot Record) that are not used by
Windows, but which can interfere with both OS/2's Logical
Volume Manager and with Linux (affecting dual boot computers
such as mine). I have read reports that C-Dilla slows down
disk reads and writes of unrelated applications. C-Dilla
was insinuated into TurboTax several months ago, causing a
firestorm of protest, malfunctioning computers, and rapid
withdrawal of C-Dilla by the publisher. How C-Dilla works
specifically with the OED, I don't know -- because OUP says
nothing about it (which strikes me as irresponsible, and not
befitting a great University Press).

During the v1 era, to the query "Can I copy the OED2 data-
file to my hard drive?", OUP famously responded:

Yes, but we don't recommend it. The data on the
CD-ROM is compressed and optimized for retrieval
off the CD-ROM. You will not notice a significant
increase in speed when retrieving from the hard drive.

Which is, of course, sheer nonsense (as hundreds of users
subsequently proved): it is *immensely* faster to retrieve
data from a hard disk than from a CD-ROM. Soon thereafter,
OUP introduced v1.13, which disallowed hard disk operation
altogether. New documents were posted on their website
claiming that the OED could not be run from the hard disk,
period -- a blatant falsehood. It seems to me that OUP is
much more concerned with protecting their intellectual
property rights than with facilitating use of the OED, no
matter how high our cost (about USD 400 in those days).
Now, of course, v3 can be installed to hard disk, but you
need to revalidate every 90 days by inserting the physical
CD-ROM (which means that if you run multiple machines, e.g.
desktops at more than one location plus travel notebooks,
you need to shlep the physical CD wherever you go -- no such
limitation with v1).

The v3 box claims "ADVANCED SEARCH CAPABILITIES". How do
these capabilities differ from v1? I see little or no

I read in newsgroups that v3 will not run under any Virtual
PC implementations (Novell, Linux, OS/2, Apple, etc). v1
ran everywhere.

The v3 capacity to "auto-search from the clipboard" is
welcome. About 25% of the time, however, these searches run
into a dead end -- a screen that says "No entries found".
Incredibly (for an HTML or SGML application), I can find no
Back button, to restore the previous screen! I'm stuck in a
cul de sac, with no keyboard shortcut enabling escape and
restoration of the last-previous lookup (whence this dead
end link originated). In v1 the previous screen is still
displayed. Am I missing something?

Similarly, I can't find any way to adjust Font Size. On a
1400x1050 pixel screen, the default font size is too small
for my old eyes. Font Size adjustments in v1 were front and
center. I hope I'm wrong, but I can't find any Font
controls whatsoever.

Each Quotation in v1 appeared on a separate line (a line
break after every quote). They were easy to scan visually.
In v3, quotations are displayed bunched together in a single
"paragraph". Much less readable and legible, in my opinion.
Indeed, if you run v1 and v3 side-by-side, v1 appears (to my
eye) much more elegant.

The Additions are billed as a major new feature of v3. In
fact, they seem to me to be very thin and insubstantial.
Most of the time, there aren't any Additions; in those rare
cases where they exist, they aren't integrated into the main
dictionary, but simply tacked on like afterthoughts.

In sum, I question whether this "upgrade" is worthwhile, or
anything more than an attempt by OUP to provide a fresh
income stream. I profoundly object to paying through the
nose and yet being treated like a potential criminal under
layers of security. Somehow I doubt that Murray, Onions, et al.
would approve. I'd enjoy very much to know what more experienced
users think.

Robert Holmgren

Reply all
Reply to author
0 new messages