CLL: I'd like to drop the formal grammars.

[Robin Lee Powell]

In the age of ubiquitous internet, I feel that the entire Formal
Grammars chapter simply doesn't belong in a printed book.

Does anyone disagree?

--
http://intelligence.org/ : Our last, best hope for a fantastic future.
.i ko na cpedu lo nu stidi vau loi jbopre .i dafsku lu na go'i li'u .e
lu go'i li'u .i ji'a go'i lu na'e go'i li'u .e lu go'i na'i li'u .e
lu no'e go'i li'u .e lu to'e go'i li'u .e lu lo mamta be do cu sofybakni li'u

[Robin Lee Powell]

On Wed, Oct 01, 2014 at 12:59:10AM -0700, Robin Lee Powell wrote:
>
> In the age of ubiquitous internet, I feel that the entire Formal
> Grammars chapter simply doesn't belong in a printed book.
>
> Does anyone disagree?

It's worth noting that doing so will require considerable re-wording
of Chapter 1, I just discovered.

[Gleki Arxokuna]

yes, remove it please.

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+un...@googlegroups.com.
To post to this group, send email to loj...@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.

[Remo Dentato]

I would suggest to leave it in for reference. If this version of CLL serves as milestones it would be better to have everything in a single place.

I hope the YACC grammar will be replaced by they PEG grammar in next version of CLL.

[Robin Lee Powell]

Reference for who? Who is actually going to type out the YACC or
the PEG from the book, or indeed make any other use of it from the
book?

[Gleki Arxokuna]

The problem mentioned is not to be ignored.

The refgram actually must give references to certain parts of PEG.

However, it doesnt do it in CLL 1.0 and I'm pretty sure it won't do it even in CLL 2.0

What is more, I have no idea how that would help lo nintadni understand Lojban better. May be first you need to master PEG before reading CLL? That'd be silly.

The most probable case would be to rephrase PEG expressions to human language and include that to CLL.

But, ke'u, that'd be extremely hard although probably necessary in distant future.

Has anyone heard of a course something like "Complete Idiot's Guide to PEG"?

I haven't.

[v4hn]

On Wed, Oct 01, 2014 at 12:59:10AM -0700, Robin Lee Powell wrote:

> In the age of ubiquitous internet, I feel that the entire Formal
> Grammars chapter simply doesn't belong in a printed book.
>
> Does anyone disagree?

Yes. Please leave it in. The book is called _C_LL and the formal
grammar is _definitely_ part of the language. I will look at it
if I'm interested in a specific parse tree or the grammar itself and I don't
have a computer with me (yes, this really does happen, even nowadays,
and it's always going to happen for as long as computers are not implanted)

I also do hope the YACC grammar will be replaced by a PEG grammar
in the next version.

If people feel really uncomfortable because there is no explanation
of the syntax of the grammar description, then we should _add_ one
instead of removing the grammar.


v4hn

[Gleki Arxokuna]

Exactly, we should add one. Who will do it? 

If no one then it's of absolutely no use and should return to the book when it gets that explanation. Otherwise it's still not Complete.

v4hn

[Remo Dentato]

Having it there would make possible to check how ther language is supposed to be parsed. I agree it's not easy and would require a deep understanding on how the grammar has been implemented, however it would serve as a basis for any possible discussion on formal grammars.

If you simply put it in a file somewhere on the 'Net, how can you guarantee that a future reader of CLL will have access to the right version?

If I remember correctly the YACC grammar had to be tweakened somehow to resolve some ambiguity that the EBNF still had.

Removing the chapter on formal grammar would only save 30-40 pages, I guess. I would prefer to continue having them in.

[Remo Dentato]

Having it there would make possible to check how ther language is supposed to be parsed. I agree it's not easy and would require a deep understanding on how the grammar has been implemented, however it would serve as a basis for any possible discussion on formal grammars.

If you simply put it in a file somewhere on the 'Net, how can you guarantee that a future reader of CLL will have access to the right version?

If I remember correctly the YACC grammar had to be tweakened somehow to resolve some ambiguity that the EBNF still had.

Removing the chapter on formal grammar would only save 30-40 pages, I guess. I would prefer to continue having them in.

[Remo Dentato]

The formal grammar is not for nintadni, of course. Is for establishing without doubt what is grammatical and what is not.

PEG is indeed simple to read. It is more difficult to create one as there might be "side effects" that are not apparent at first sight (i.e. indirect left recursion or undetected "prefix").

I think that whatever changes is proposed to the language, the first step is to evaluate the impact on the formal grammar. Especially for any experimental cmavo.

[v4hn]

On Wed, Oct 01, 2014 at 03:48:35PM +0400, Gleki Arxokuna wrote:
> 2014-10-01 15:23 GMT+04:00 v4hn <m...@v4hn.de>:
> > On Wed, Oct 01, 2014 at 12:59:10AM -0700, Robin Lee Powell wrote:
> > > In the age of ubiquitous internet, I feel that the entire Formal
> > > Grammars chapter simply doesn't belong in a printed book.
> > >
> > > Does anyone disagree?
> >
> > Yes. Please leave it in. The book is called _C_LL and the formal

> > grammar is _definitely_ part of the language. [...]

> >
> > I also do hope the YACC grammar will be replaced by a PEG grammar
> > in the next version.

So, I just had another look at the relevant chapter.
pe'i It is, indeed, important to keep the formal grammar within the book.
However, that does not mean we have to have two representations of the very same
thing in there. The EBNF grammar should be enough, so I agree that (by pages)
most of the "formal grammars" chapter (i.e. the yacc grammar) could
(and probably should) be removed. This would include all the rule numbering
in the EBNF grammar that references the yacc rules.

> > If people feel really uncomfortable because there is no explanation
> > of the syntax of the grammar description, then we should _add_ one
> > instead of removing the grammar.
>
> Exactly, we should add one. Who will do it?
> If no one then it's of absolutely no use and should return to the book when
> it gets that explanation. Otherwise it's still not Complete.

Turns out the EBNF rules are explained and even described as "human readable".
So there's no problem here.


v4hn

[mukti]

On Wednesday, October 1, 2014 4:59:12 AM UTC-3, Robin Powell wrote:

In the age of ubiquitous internet, I feel that the entire Formal
Grammars chapter simply doesn't belong in a printed book.
Does anyone disagree?

At more than 100 pages (more than 10% of page count) in the current build, it adds a lot weight but does not offer proportionate value. Which is to say, I agree. It's not likely to serve readers who aren't conversant in formal grammars, whereas those who are familiar with such things may not find the print format readable: I know I don't.

I agree with v4hn that the EBNF grammar is worth more consideration than the YACC grammar, but mostly because at only 20 pages it is less burdensome, and not because I think print is ultimately a dignified or useful format for it.

mi'e la mukti mu'o

[la durka]

FWIW, from IRC:

8:53 <Broca> How close are we to having a CLL v1.1 ready for sending to the printers?
8:53 <Broca> I might be able to tweak the formal grammar chapter to have a less insane page-count, but probably not until this weekend.

[TR NS]

On Wednesday, October 1, 2014 3:59:12 AM UTC-4, Robin Powell wrote:

In the age of ubiquitous internet, I feel that the entire Formal
Grammars chapter simply doesn't belong in a printed book.

Does anyone disagree?

My opinion, I'd be more concerned with adding a basic lexicon, so if page count is an issue, then bye-bye formal grammar.

[Alexander Kozhevnikov]

On Wed, 1 Oct 2014, Robin Lee Powell wrote:
> In the age of ubiquitous internet, I feel that the entire Formal
> Grammars chapter simply doesn't belong in a printed book.

s/ubiquitous/ubiquitous for some large portions of society but definitely
not all, and not everywhere, and often obscenely throttled or otherwise
limited for many/

^ I think that is more representative of the truth.

I don't have an opinion about removing or keeping the chapter one way or
the other at this time (insufficient experience in using the CLL to learn
so far), but I hope that the above is kept in mind either way. There are
still plenty of people even in developed countries like the US who might
have access to a book, but not to internet (at least not at the time when
they might want to look it up, and not internet that isn't too expensive
or slow to justify looking up and downloading the grammar).

Regards,
mu'o mi'e .aleksandr.kojevnikov. do'u

[Craig Daniel]

The book is commonly used as a tool by nintadni, for whom the formal grammars aren't helpful in the last - but pe'i the existence of a formal grammar is part of what makes Lojban Lojban, and the primary purpose of the CLL is to be a reference grammar. I like the suggestion of keeping only the shorter of the two, and updating to the PEG in v2.0, but i ultimately see the English text as an explanation of and supplement to the formal grammar rather than vice versa.

- mi'e .kreig.

[Bob LeChevalier, President and Founder - LLG]

I hate to disagree, but I learned Lojban grammar by writing the YACC
grammar, and I have used it as a reference pretty much ever since. I do
not find the EBNF useful, because in the interest of conciseness, it has
fairly complex structures on the right side; I get lost trying to take
apart a construct. The YACC typically adds one piece at a time,
possibly with a terminator, and for some reason, the simplicity of each
step has always meant that it has been easy for me to understand.

(We included both in CLL because some found the EBNF easier to
understand, whereas others found the YACC easier. Neither Nora nor I
have any clue how to use a PEG grammar. But if we had had one, we
probably would have included it too).

lojbab

[Spacenut]

As a layperson, not knowing anything about the formal grammar, I say we absolutely should keep it in there. Though it may be of no use to me, The Complete Lojban Language is not complete without the full grammar. Also, as a less technical and more psychological argument, it reinforces the idea that Lojban is 100% syntactically unambiguous to people who don't know anything about the parser. The fact that it can be interpreted (don't know if that's the right word there) by a computer is one of the reasons why I'm learning Lojban, and having tangible evidence of that is pretty powerful.

On Wednesday, October 1, 2014 3:59:12 AM UTC-4, Robin Powell wrote:

[Alex Burka]

I don't have much of an opinion since I prefer the PEG and obviously that's not going in this edition, but surely Lojban is still confirmed as machine-parseable by the inclusion of one and not both grammars. And I don't mean to deny the usefulness of the YACC grammar to Bob, but I'll point out that the fact that it was pedagogically useful to _write_ doesn't necessarily imply that it's pedagogically useful for others to _read_ %^)

mu'o mi'e la durka

--
You received this message because you are subscribed to a topic in the Google Groups "lojban" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lojban/KpcylrlDNX0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lojban+un...@googlegroups.com.

[Jonathan Jones]

I'm sure there's a reason why we can't just release both a with and without version, see which one gets purchased more, and use that info. to determine future version content?

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+un...@googlegroups.com.

To post to this group, send email to loj...@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.

--
mu'o mi'e .aionys.

.i.e'ucai ko cmima lo pilno be denpa bu .i doi.luk. mi patfu do zo'o
(Come to the Dot Side! Luke, I am your father. :D )

[TR NS]

On Thursday, October 2, 2014 12:00:15 AM UTC-4, aionys wrote:

I'm sure there's a reason why we can't just release both a with and without version, see which one gets purchased more, and use that info. to determine future version content?

What about having two volumes. The second could be a dictionary plus formal grammars. Perhaps also the catalogue of slema'o, and personally I'd like have a separate cmavo glossary.

In other words, the second volume is reference material.

[Jacob Errington]

On 2 October 2014 05:46, TR NS <tran...@gmail.com> wrote

What about having two volumes. The second could be a dictionary plus formal grammars. Perhaps also the catalogue of slema'o, and personally I'd like have a separate cmavo glossary.

In other words, the second volume is reference material.

I like this idea very much. Including the formal grammars as an appendix to the dictionary makes more sense to me.

.i mi'e la tsani mu'o

[Robert Slaughter]

There was supposed to be a printed dictionary shortly after the publication of CLL 1.0, very similar to this. That worked out well.

--

[Michael Turniansky]

  I agree with you, Robin.  I don't believe the formal grammar belongs in the printed book, or at least not _this_ printed book.  It might be fine as a separately printed  appendix.  I, for one, have gotten along fine without it, never really gave it more than a glance, and my eyes glaze over whenever other people start referencing it on the email list.  It's not that I don't understand formal grammars.  I do have a CompSci degree.  But I believe that chapter is only of interest to those people who are  going to be programming in it, with it, or about it (e.g. camxes, jbofi'e, and cakyrespa)

   It's no different in my business.  For example, there are user manuals for printers, but I usually need something deeper, the prorammer's manual, because I need to learn how the down-and-dirty nitty-gritty works in order to get them to work with our applications.  99.99% of the public doesn't need that.  Same with lojban.  I didn't (And don't) need that chapter to speak lojban, nor does anyone else (Except, apparently, lojbab?).  Dump it, and make it available on request.

                --gejyspa, jumping out of my 3-month email backlog just for you, camgusmis <3

[Robin Lee Powell]

Including it means I have to do the work to maake it look decent on
the page, which is the work I don't want to do.

[Robin Lee Powell]

Oh, I hadn't thought of dropping only the YACC; that's a very
interesting idea.

> > To unsubscribe from this group and all its topics, send an email to lojban+un...@googlegroups.com (mailto:lojban+un...@googlegroups.com).
> > To post to this group, send email to loj...@googlegroups.com (mailto:loj...@googlegroups.com).

> > Visit this group at http://groups.google.com/group/lojban.
> > For more options, visit https://groups.google.com/d/optout.
>

> --

> You received this message because you are subscribed to the Google Groups "lojban" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to lojban+un...@googlegroups.com.

> To post to this group, send email to loj...@googlegroups.com.
> Visit this group at http://groups.google.com/group/lojban.
> For more options, visit https://groups.google.com/d/optout.

[TR NS]

On Friday, October 3, 2014 10:39:49 PM UTC-4, Robin Powell wrote:

Oh, I hadn't thought of dropping only the YACC; that's a very
interesting idea.

You don't have to (which will make lojbab happy) if you make two volumes. I especially like that idea because then I can be reading the first volume while referencing the second. 

[Robin Lee Powell]

Again: what I wanted to avoid was the work involved in making the
YACC actually look good on paper.

[MorphemeAddict]

If you take out the grammar with the intention of putting it in a second volume, then the work for formatting that grammar can be done separately, even postponed, perhaps indefinitely. If the demand for it isn't very high, then it might not ever get printed, but that's not such a bad thing. 

stevo

[Gleki Arxokuna]

I don think one should postpone publishing CLL 1.1 because of the requirement of YACC.

I'm sure many people are ready to buy CLL without YACC and buy the second volume later.

[Jey Kottalam]

IMHO: Function over form; completeness over prettiness.