JACKIE CHAMBERLAIN'S SURVEY: 1997

From:   	Jackie Chamberlain [jchamberlain@pe.com]
Sent:   	Thursday, September 18, 1997 4:41 PM
To:     	NewsLib
Subject:        Results of survey on corrections of Web editions

Last week I posted a question about corrections on Web editions which set
off a far more debate than I expected. I wasn't terribly surprised by what
I heard about correcting Web editions -- and it certainly confirmed what
Nora Paul said in her piece in the April 1996 Searcher: ("The Times, They
Are A-Changin': Newspapers on the Web (Part One --The New York Times, ")
especially about corrections. 

The fact is, *there are no standards* for Web editions. What did surprise
me was that the discussion quickly segued into one about corrections of
traditional news databases, and it's apparent large cracks are appearing
in the standards there. 


Eight papers provided me with substantive information -- most of them
privately. Here are the results. This is hardly scientific and there's
some overlap because some do one thing or another depending on the
circumstances, nature of the error, cooperation with the newsroom, etc.
Since so many responses were private, I respect that and the sources are
not identified. 


Of the 8 (and in a few cases the references seem to be to traditional
databases too): 


1. Include corrections as a correction file and/or as freestanding stories
    and/or in index = 6 (2 don't say) 

2.Append correction and flag it to draw attention = 2

3.Append correction and change error in text, replacing 
   original = 2

4.Change error in text and note a change was made  = 2

5.Change error in text but only append correction if it's significant
   (i.e. an actionable error) = 3

6.Make no corrections but include a corrections file = 2

7.Kill offending story rather than correct it = 1

8. Who does it? Web editor/production  = 7
    Unknown = 1
    Library = 0 (except in 1 case where Web edition taken from   
      the Library database and in that case the Library has nothing
      directly to do with Web.


Paper 1. Treat Web edition same as in-house archives. Publish correction
as separate news item. Append correction at top of story, framed with
flashing "icons" to catch the eye. Never append at bottom, always at top.
Web edition is actually traditional archives loaded onto Web. 


Paper 2. If correction comes while story still on main site (24 hrs)
re-edit story to include the correction and run only the correct story.
Library not directly involved. 


Paper 3. If significant error, change error ASAP with correction also
appended, but no notation that error was also fixed. Run major corrections
that paper runs via file in Web news index unless offending story wasn't
in Web edition. Have difficulty picking up corrections (this from Web
editor) from local editions because bureaus don't always put them on
budget. Not always aware of need for corrections or find out too late
because newsroom fails to inform. Difficult to manually fix stories. Run
all corrections to anything that has been online, even if story has
dropped off. Library not involved. 

Paper 4. Web edition separate operation. Webbers don't bother attaching
corrections but do load corrections as separate stories. Web stories stay
online for a week. Library has no role in Web edition

Paper 5. One of minority that advocates changing errors not only on Web
edition but in newspaper archive, with notation that printed edition
contained error that was corrected for electronic file and cite where the
correct, a free-standing file, can be found. Consider microfilm, not
electronic archive, official record [but of course there's no film record
of Web edition]. But do append significant corrections, and do not
advocate extensive "rewriting of stories." Actually only change errors
such as misspelled names, wrong phone numbers, wrong job titles.... Not
clear what role Library plays in Web edition. 

Paper 6. Make correction in Web versions of stories and note which stories
were "updated." If story no longer online, run separate correction in
daily index, noting date of offending story and section in which it
appeared. Correction usually late because have small Web staff and news
desk not yet in habit of keeping Web editor informed. Library has no role
in Web edition. 

Paper 7. Post corrections as separate "stories," but make no effort to
link correction to original (even though they have a two-year archive of
the Web edition online --distinct from library archive). In case of
egregious errors with lawsuit potential, Library will alert Webbers and
suggest they either attach correction or kill original. Webbers always
just kill original. This has happened about 6 times in 2 years. Library
has no direct role in Web edition. 

Paper 8. Same person apparently involved in maintenance both of newspaper
database and Web edition (though not entirely clear). Advocates changing
errors in text (of both database and Web edition) but also feels attaching
correction is important. Feels mi nor errors should just be changed
(spelling, misidentification). The need to "rewrite should involve a
reporter/editor" if, and only if, the correction is published. [I had
asked who's supposed to do the rewrite when the error was the result of
lack of clarity in writing]. The Library is not involved. 


So there it is. 

Two conclusions from the responses I received privately and the debate on
NewsLib: 

	(1)There are no standards for identifying, correcting, and
editing the errors in newspaper Web editions. Further, it's no doubt
significant that the Library staff isn't involved in any of these Web
editions (though at least one is a librarian by training.) 

	(2) There are shortcomings in quality control standards for
electronic library archives of newspapers. How significant are they? How
far can these lapses continue before they substantively effect the quality
of the newspaper, its archives, and indeed the foundation of journalistic
accuracy? 

Exactly where is this discussion heading? We have yet to find out, but we
do know this much: We all need to make the effort to find out. Perhaps
we need to keep a few fundamentals in mind. A direction can be found in
this quote from the Society of Professional Journalists' Code of Ethics: 
 
		"Test the accuracy of information from all sources
		and exercise care to avoid inadvertent error."

And this quote, from news librarian Anne Mintz, Forbes, Inc. director of
information services in her article, "Quality Control and the Zen of
Database Production," published in Online, Nov. 1990, pages 15-23, has
special merit in this context:

		"Full-text follies: . . . by highlighting these
		gaps and traps (errors and inconsistencies) to database
		providers and online searchers, progress can be made in
		improving the overall products and services of online
		full-text newspaper databases."

If this is the beginning, what is the next step for those of us involved
with the accuracy, integrity, and reliability of newspapers? Do we have
roles to play on the Web editions of our newspapers the way we do
(should?) with print and online database editions? If we do, what is it? 
The discussion continues.

Jackie Chamberlain               jchamber@pe.com
The Press-Enterprise             V - (909) 782-7579
Riverside CA                     F - (909)  782-7572