- Home
- Get A Printable Dictionary
- Search Best Words
- Recent Changes
- How You Can Help
- valsi - All
- valsi - Preferred Only
- natlang - All
- natlang - Preferred Only
- Languages
- XML Export
- user Listing
- Report Bugs
- Utilities
- Status
- Help
- Admin Request
- Create Account
|
Discussion of "poiklo"
Comment #1:
Offset?
|
Curtis W Franks (Sun Dec 7 01:55:53 2014)
|
Subsequences just have an extra index/map (internally composed). I do not understand this definition.
|
-
Comment #2:
Re: Offset?
|
Ilmen (Sun Dec 7 12:26:48 2014)
|
krtisfranks wrote: > Subsequences just have an extra index/map (internally composed). I do not > understand this definition.
coi la .krtisfranks.
Actually I didn't know that "subsequence" was a technical term different in meaning from "substring"; what I intended to mean was "substring". I will correct the definition accordingly.
Making lujvo for technical terms is not always easy, as the place structure is somewhat constrained.
What do you think of the following definition for poiklo: «x1 (sequence/string) has an occurrence at position/offset x2 (number) in sequence/string x3»
Examples: «zo ti poiklo li vo (lo lerpoi pe) lu mi ti do dunda li'u», «zo ti poiklo ci da (lo lerpoi pe) lu ti titranti li'u» (I'm not quite sure whether quotes can be treated directly as phoneme sequences, but this seems sensible enough.)
mi'e la .ilmen. mu'o
|
-
Comment #3:
Re: Offset?
|
Wuzzy (Sun Dec 7 14:53:07 2014)
|
> Examples: «zo ti poiklo li vo (lo lerpoi pe) lu mi ti do dunda li'u», > «zo ti poiklo ci da (lo lerpoi pe) lu ti titranti li'u» > (I'm not quite sure whether quotes can be treated directly as phoneme > sequences, but this seems sensible enough.)
A string is an arbitrary sequence of characters (see lerpoi). The closest equivalent in Lojban for this are zoi-quotes, not lu-quotes. Your “lu” seems a bit contrived, since spacing in Lojban is a bit liberal. Thus, “lu mi ti do dunda li'u” is (and should be) treated as identical to “lu mitido dunda li'u”. Unless you invent a rule to count positions in “lu” strings, I suggest to avoid “lu” and “lo'u” quotes as sumti for poiklo.
I think that poiklo is most useful with zoi-quotes.
I guess a more practical example would be something like this: “zoi gy.http.gy. poiklo li pa zoi gy.http://jbovlaste.lojban.org/.gy.” → “‘http’ is a substring at position 1 of the string ‘http://jbovlaste.lojban.org/’.”
Oh, and that counting begins by 0 is not at all a “standard” in programming languages. There are a couple of languages where counting starts by 1. I have removed that part of the sentence from the notes.
|
-
Comment #4:
Re: Offset?
|
Ilmen (Sun Dec 7 22:20:11 2014)
|
Wuzzy wrote: > A string is an arbitrary sequence of characters (see lerpoi). The > closest equivalent in Lojban for this are zoi-quotes, not lu-quotes. > Your “lu” seems a bit contrived, since spacing in Lojban is a bit > liberal. Thus, “lu mi ti do dunda li'u” is (and should be) treated as > identical to “lu mitido dunda li'u”. Unless you invent a rule to count > positions in “lu” strings, I suggest to avoid “lu” and “lo'u” > quotes as sumti for poiklo. > > I think that poiklo is most useful with zoi-quotes.
I think lu-quotes are either lerpoi (phoneme sequences), or vlapoi (word sequences). If they are lerpoi, there needs to be a default spacing, probably either fully-spaced Lojban or spaceless Lojban, so that «lu mi ti dunda do li'u» is considered as identical to «lumitidúndadoli'u». If the default is spaceless Lojban, then no matter how many space do you put in the lu-quote, the resulting lerpoi would contain spaceless Lojban. If lu-quotes are vlapoi, there's no spacing issue though, because each word is in a different slot in the sequence.
|
-
|
Comment #5:
Re: Offset?
|
Curtis W Franks (Tue Dec 9 06:57:10 2014)
|
Wuzzy wrote: > > Examples: «zo ti poiklo li vo (lo lerpoi pe) lu mi ti do dunda li'u», > > «zo ti poiklo ci da (lo lerpoi pe) lu ti titranti li'u» > > (I'm not quite sure whether quotes can be treated directly as phoneme > > sequences, but this seems sensible enough.) > > > A string is an arbitrary sequence of characters (see lerpoi). The > closest equivalent in Lojban for this are zoi-quotes, not lu-quotes. > Your “lu” seems a bit contrived, since spacing in Lojban is a bit > liberal. Thus, “lu mi ti do dunda li'u” is (and should be) treated as > identical to “lu mitido dunda li'u”. Unless you invent a rule to count > positions in “lu” strings, I suggest to avoid “lu” and “lo'u” > quotes as sumti for poiklo. > > I think that poiklo is most useful with zoi-quotes. > > I guess a more practical example would be something like this: > “zoi gy.http.gy. poiklo li pa zoi gy.http://jbovlaste.lojban.org/.gy.” > → “‘http’ is a substring at position 1 of the string > ‘http://jbovlaste.lojban.org/’.” > > Oh, and that counting begins by 0 is not at all a “standard” in > programming languages. There are a couple of languages where counting > starts by 1. I have removed that part of the sentence from the notes.
I have made a proposal to la gleki about something quite closely related. I decided that spaces do count between brivla because they must be encoded somehow (alternatives include marking vowels or syllables in some way): they convey indestructible information in that context. In that case, strings have an extra "hidden letter", where I interpret each letter to be a unit of conveyed information (meaningful in context). The information is what is being counted, actually, not symbols or phonemes. I also start counting from 1 in Lojban, but we should really formalize and systematize it as a community after much debate. I usually specify at least the first element in my definitions and also try to allude to the ordering. I will try to find the message and post its content somewhere. This has combinatorial uses, btw!
I much prefer the current definition and do not have any immediately obvious concerns.
|
-
|
|
|
|
|