RE: Multilingual Documents [was: HTML forms and UTF-8]

From: Chris Pratley (chrispr@microsoft.com)
Date: Tue Nov 23 1999 - 21:08:15 EST


Yes, our research covered web documents (not sure about the Xerox or
Progress work). I will note a few more things on this topic:

1. At some point, it is easier to just globalize the entire product and make
it completely, unabashedly multilingual rather than keep hacking away trying
to get specific combinations to work. We reached that point a couple of
versions ago in Office, and as a result we have a nice clean globalized,
multilingual Office2000 which is much easier to maintain and add language
support to than our old model of different builds and even different source
for different languages. As Erik notes, Mozilla is doing the same thing for
Navigator 5.0.
2. If you are thinking about documents, it is easy to remember the
multilingual ones, but the many letters, memos, plans, and emails that are
monolingual seem to slip our memory. It is those that drive the ratio up.
3. It is easy to think of plenty of anecdotal evidence for multilingual
documents, but ultimately it is the statistical evidence that is more
important. And as Peter notes, it needs to be stats taken on people who
could easily have created multilingual docs if they wanted to.
4. It is simply easier for people to create monolingual documents. Even if
the author is fluent in two languages, keeping the versions in sync with
edits and making sure they say substantially the same thing is a hassle most
people don't want (but many governments are forced to adhere to)
5. It is easy to confuse "documents" (which I am using as a generic term
covering "collection of related phrases or sentences", ranging from e-mail
to formal papers) with "publications", which are more formal and more likely
to be multilingual especially in some countries, but which are a tiny
fraction of documents. (When I was growing up in Quebec, I did not see a
large fraction of documents that were bilingual, except for certain
government publications and advertisements)
6. Another point made by a couple of people is that you can measure the
ratio of documents (as we were discussing), or you can also measure the
percent of users who *ever* have to create even one multi-language document.
This percentage is considerably higher than the percentage of multilingual
documents, more like 10-30% of users as opposed to <1% of documents. If you
are strictly looking at business cases, then this is a misleading number,
because a feature that makes a significant percentage of users' work easier
is more valuable than one that makes a tiny percent easier. So measuring
per-document makes more sense.

And, just to reiterate, anyone who knows me knows that I (and Microsoft) are
all in favour of supporting multilingual usage in Office (and Windows, and
other applications), and have a track record to prove it. I'm just trying to
point out some realities.

Chris Pratley
Lead Program Manager
Microsoft Office

-----Original Message-----
From: peter_constable@sil.org [mailto:peter_constable@sil.org]
Sent: November 22, 1999 10:08 PM
To: Unicode List
Subject: Re: Multilingual Documents [was: HTML forms and UTF-8]

       Well, I can't argue when people tell me that three companies
       did reasonably extensive research and found the same results.
       But that's definitely a different argument than what was
       originally presented: "Multilingual documents are rare,
       therefore people must not want them." There was a logical
       fallicy there I wanted to point out, but Chris, Joe and Tex
       have provided evidence that supports the conclusion.

       I wonder if the research done included web documents, which
       was, after all, what started this discussion: difficulties in
       creating multilingual web documents. I can easily imagine
       *lots* of potential need for multilingual web documents, and
       others have suggested the need. For example, I find it not
       uncommon to perform a web search, searching on a monolingual
       English string, and to get result pages with multilingual - and
       multiscript - content. I've come to expect that the portion in
       another script will be illegible gibberish (though IE 5 is
       definitely a move in the right direction). I think it's a valid
       conclusion - and, if I recall, this was the original point of
       this aspect of this thread - that there is a real market need
       to improve Netscape Navigator's ability to deal with
       multilingual documents (and forms).

       There are also the minority who have a need for multilingual
       (web and traditional) documents, who are part of the
       developers' market. We may not represent enough of the revenue
       to sway any business model, but we still hope technology will
       advance to provide the capabilities we'd like to have. Whether
       that comes through explicit design or as a side benefit of a
       program of software globalisation doesn't really matter, as
       long as the desired capabilities arrive.

       Peter

       From: <texin@progress.com> AT Internet on 11/22/99 04:16 PM CST

       Received on: 11/22/99

       To: Peter Constable/IntlAdmin/WCT, <unicode@unicode.org> AT
             Internet@Ccmail
       cc:
       Subject: Re: Multilingual Documents [was: HTML forms and
             UTF-8]



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:56 EDT