[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lojban] Lojban in 12 weeks?
At 08:02 PM 10/3/02 +1000, Nick Nicholas wrote:
The Lojban brochure includes the claim: "A working vocabulary including
the complete set of 1350 root words takes about 8.12 weeks of study at
1 hour per day."
Robin.CA's apophthegmatic response on the Wiki errata page: "Bullshit".
I have only gotten a couple of logs from Lojbanists. I don't have Robin's
so far as I can tell.
One person sent a log spread out over 9 1/2 months. There were 32 sessions
with lots of gaps of weeks or more, and then the person started over. He
then finished the new words in 62 sessions after 2 months, and then
continued to record data until he reached 142 sessions after 4 months, when
it looked like he was averaging around 85% on his Under control words. His
lessons appeared to take a little more than an hour while he had new words,
and then dropped to around 45 minutes. Probably 120 hours after he restarted.
Another person did 75 sessions of around 20 minutes over 9 weeks to get
through the new words, but that was in the beginner mode with no recall
lessons. He never sent a log of Gaining control mode.
A well-known non-English-native Lojbanist who had been involved in the
language for a couple of years before starting Logflash took 73 sessions of
a half hour to get through the new words, and was near 100% by session
90. He quit recording after 120 sessions, a total of 8 days (that is
called a Lojban intensive course %^). Probably 45-60 hours total.
Another non-English native Lojbanist had around 11 sessions intermittently
over several months, then got really serious (but didn't reset the words)
doing some 50 sessions each 15 minutes long or so over a Christmas
break. He then slowed down, doing 80 more sessions (still 10-15 minutes
long) over the next 6 months before he finished the New words. He was at
near 100% level at that point, though he continued another month before
going into Maintenance mode. In intermittent lessons over the next year
and a half, he typically had 80-90% correct. Probably around 45 hours
total before he went into maintenance mode.
I would say that 8 weeks (60 hours) is certainly achievable for those who
use the program rigorously, and who start in "Gaining Control" mode which
unfortunately has not been the default for a while (the recognition-only
beginners lessons appeared to add 2 months to the total time, albeit with
less than a half hour per day - this spreads out the effort, but probably
makes it more possible to get through the words more intensively when
Gaining Control is used. But we have no data to support this.
However, it appears that relatively few people use the program rigorously
as it was designed, and rather than an hour a day for 8-12 weeks, either do
it intensively in which case it may take fewer hours but a lot more than an
hour a day, or intermittently. I suspect the 120 hour figure for the first
person would reflect the most typical result, which would suggest that we
should say "8-16 weeks" to cover the range of reported experience (it has
been far too long for us but I think Nora and I took around 90-100 sessions
in 1987-8 - hers a little less than an hour, mine an hour and a half until
I got through the new words and then a half hour for an overall average of
an hour. Tommy Whitlock was averaging closer to 2 hours per session,
suffering particularly badly because he occasionally skipped 2-3 days,
which seriously raises the error rate the first time through. None of us
was actually doing anything with the language then because we didn't have
much of a grammar developed, so this was pure rote drill).
If anyone else has used LogFlash extensively (ideally having gone through
all the new words and preferably up to maintenance mode) and has not sent
me logs, the files are need are the two files labelled ".fl1" with your
name in the filename. Thorough analysis of the logs takes a while
including a software program that I don't remember how to use right now,
but the above numbers came from a quick scan (since I have a pretty good
idea what to look for).
lojbab
--
lojbab lojbab@lojban.org
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA 703-385-0273
Artificial language Loglan/Lojban: http://www.lojban.org