wwei47 wrote: ↑March 5th, 2021, 8:53 am
Moosey wrote: ↑March 5th, 2021, 8:47 am
I suggest we start a thread for an SL nomenclature project.
Hope that goes well, see this post:
viewtopic.php?f=2&t=1452&hilit=siamese& ... 625#p49146
The point is that nomenclature is difficult and might lead to things like "(((Long long long long bookend) weld (Long bookend siamese (Block on table))) on ((Long long long long bookend) weld (Long bookend siamese (Block on table)))).
Hat siamese (Beehive siamese beehive siamese beehive siamese beehive) siamese (Hat siamese (Table on table))".
Yup -- anyone who is tempted to start a new nomenclature thread should please try reading through the
last several nomenclature threads first.
The thing is, it's kind of fun to
write great long strings of text describing complex still lifes. But then it's no fun at all to
read them, after they've been written, or compare one piece of descriptive nomenclature to another to see if they're the same object. Nobody would want to memorize the hundreds of micro-rules you'd need to know to be able to name and/or reconstruct even a moderate-sized still life reliably -- no matter what specific conventions you might decide to use.
(It would be much easier to just learn the relatively small amount of detail you need to know to reconstruct apgcodes -- that's mostly just memorizing bit patterns for base-32 characters.)
Hunting wrote: ↑March 5th, 2021, 8:10 am
MathAndCode wrote: ↑March 4th, 2021, 7:26 pm
To clarify some nome[n]clature discussed in Discord for the fast components collection, this is an up cis-carrier siamese bookend...
This is a down cis-carrier siamese bookend...
This is an up trans-carrier siamese bookend...
This is a down trans-carrier siamese bookend...
^This is how dire the SL nomenclature currently is. Even the fundamental parts needs to be clarified...
Speaking for myself here, a clarification like the above doesn't seem to clarify anything at all. It doesn't explain clearly _why_ "up", "down", "cis", and "trans" are applied in these four cases. Is there a canonical order to the two parts joined by "siamese" -- like why isn't it "bookend siamese [up|down] [cis|trans]-carrier", in alphabetical order? Is there a canonical orientation? i.e., if I mirror the pattern across the X axis, do "up" and "down" change? etc.
It only takes very slightly more complex patterns than these to illustrate how hopeless it is to come up with unambiguous and widely agreed-on canonical descriptive names. Why is this thing
Code: Select all
x = 7, y = 7, rule = B3/S23
5b2o$4bobo$4bo$2bobo$bobo$bo$2o!
a
fuse with tail and integral and not a "
melusine weld eater tail", or "eater head weld eater tail", or some other equally horrible description? Add a few more bits and a few more overlaps, and you very soon start seeing descriptions that are really too ambiguous to allow the object to be reconstructed reliably from its component pieces.