[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: How many levels

On Thu, Sep 3, 2009 at 2:07 PM, Matt Arnold<matt.mattarn@gmail.com> wrote:
>
> xorxes,
>
> Your plan looks acceptable to me, with a couple of qualifications.
>
> If I take level 500 and pass it, am I then called a "level 500"?

Maybe you'd be called a munonova'e.

I don't think the name is all that important, I'm only suggesting that
we use some naming scheme that allows for inserting more intermediate
levels when/if they become necessary, without having to rename already
established tests.

> What happens if we add a level 100* later? Someone who comes along
> later has to pay more dues than I did to get to the same level? Or
> would we charge me a test fee for 100* when we make it available, but
> I don't have to take it? Neither of those seems quite right... did you
> have something in mind? That should be handled delicately.

I think one should be free to start at whatever level one chooses, or
skip levels. If someone feels they can pass the level 500, they should
not be forced to pass the level 100 first if they don't want to. Or do
you want people who already know that a level 100 will be trivial for
them to have to take it anyway?

> However, it seems easier to make a simpler test than to make a more
> advanced one. Personally, I see no reason why not to start from the
> earliest level we consider meaningful, and obviate the above
> questions. :)

I don't know if it's easier to make a test when you are constrained to
use a small set of words. You have to be more inventive with the
questions. But I'm not opposed to starting at level 100.

> We don't have to work on more than one or two levels at a time. We can
> just start at the level 100* test, and work on the next higher only
> when we're ready.

I've been looking at the word frequency lists in
http://www.lojban.org/tiki/Word+frequency+lists

I think we should count cmavo and gismu separately, because of the 100
most frequent words only 18 are gismu and the rest are cmavo. I don't
think a test based on that makes a lot of sense. Maybe a breakdown
like 40 cmavo - 60 gismu would make more sense.

mu'o mi'e xorxes