Re: [lojban] Re: CLL and modern Lojban

[sukender1@gmail.com]

Hm... good point. Are you absolutely sure the community can't self-organize ?

You just witnessed it with your own eyes (PEG issue). It's only you who must study the subject, not delegate dealing with it to someone else.

Sorry I don't understand clearly your answers.

You saw how someone thought that PEG grammar can resolve parsing ambiguity, which is wrong.

This wrong opinion can proliferate in the community of people who don't understand these core parts of Lojban, which can result in deteriorating of Lojban and in the long run abandoing it. It's easy to introduce bugs than to support a functional program.

Agreed about bugs introduction.

(Thanks for rephrasing and explaining)

Okay I understand you think no self-organization can happen. What do you mean with the second sentence? What must I study, according to you?

Lojban grammar, Lojban language and its underlying principles. In the case of PEG you can study CFG, YACC, BNF, PEG, probably everything from regular expressions to turing, machines, how they work, you may stuydy the history of PEG grammar of Lojban (logs here).

Understood. I am currently self-teaching many things, including some specific language aspects, logic, machine learning, and more (aside from Lojban itself). This is a lot of work, of course. So let me add some precisions:

• Most people generally can't self-evaluate accurately enough, because they don't estimate what they don't know. So you're right about saying one should study before proposing breaking changes. But on the other extreme, one just can't wait for "the ultimate expert level". Or else only 2-3 people will be able to perform changes. I'm convinced there is a good balance in-between (closer to "expert" than "beginner" IMO, but not at any extreme point).
  
• This is why I only give proposals about what I feel comfortable with (= expert-enough).
• Thus, I personally don't want to enter that PEG/YACC/other debate. I don't feel legitimate here, as for many other topics. And this is not a topic I want to tackle.
• Some topics cover multiple domains (say "language, logic, and parsers" for instance). Except for specialists & professionals, we can't expect people to handle all fields well. This is why cooperation is mandatory (IMO, here also).
• I personally don't tell people what to do, or what "could be nice" without doing it by myself too. This is exactly what I'm currently doing: I collect info and discuss, while I'm preparing stuff to propose.
• I can't tell for others. I'm not the "others".
• This makes me dislike reading things like "we should have..." without anything proposed.
  
• Yes, proposing things is hard work, and some work may simply be thrown away because others reject it. People should accept this fact and propose notwithstanding... or never say "we should have" sentences.
• I really like analysis messages, telling me what could be good and what could be wrong. But I would also like (from all) to not just tell "Not ok", but "Not okay BECAUSE blah-blah-blah, and MY PROPOSAL IS blah-blah-blah". I like constructive criticism.
• You gave me a link to the discussion forum, so I have access to anything. Right. But as for many others (I guess), my spare time is very limited and I just can't afford to read all through those conversations. Unfortunately, it seems there are no summary about some hot topics. Of course I blame no one, because summarizing also takes time. And it may induce a subjective bias. I have no solution on that point yet. But I must admit your remarks and fixes do help.

 

In that case, any better idea to avoid current issues? I mean:

1. Actual "experts" may be lead to an even more cluttered situation.
2. The community is heterogeneous and cannot tell who are to be the "meritant" (worthy) ones.

 

Or maybe some kind of algorithm checking the "merit" of the user by scanning cor contributions and posted messages? Well, that last idea may be cumbersome, but...

And about this? May we find a set of rules that can roughly "sort" user in a few groups? Maybe that would be imprecise and imperfect, but that may be better than nothing.

What are the problems you currently have? 

The problem is about grabbing points of view on how to define "expert" or "merit" in the topic. You told me we can't rely on current "experts", and that sounds fine to me. Do you have anything to propose, regarding the submission protocol draft I posted?

Why do you even need experts? What's your ultimate goal?

• There are plenty of proposals everywhere. This is a fact, I can't just ignore this. Even the errata and very basic bug fixes are still "proposals", as there is no official "CLL 1.x" containing them. The status of CLL 1.2 is just unclear.
• I would like proposals to be organized. Having unregulated stuff everywhere leads to heterogeneous proposals, some being interesting, and some being (as you said) very prone to break things and introduce bugs.
• Organizing requires experts (or an approaching concept), to review, evaluate, discuss, etc., and finally avoid bugs.
  
• There may be language experts, logic experts, technical experts, etc.
• This also requires tools & procedures. I'm working on some, to be submitted here soon.
• My "ultimate goal" is twofold: I want to help increasing the number of Lojbanists, and to help providing others what I could not find when arriving, that is to say a clear situation.
• Those goals are entangled: I can't promote Lojban if the situation isn't clear. Actually, I stopped promoting Lojban because of this. More than this, I'm a bit ashamed of having promoted it, now I understand how far the situation is cluttered. I had never imagined this to be even possible, hence my former confidence about promotion (once again, I blame no one).
• Increasing the number of Lojbanists is obviously good for anyone, isn't it? Notoriety will help getting widespread support, more native texts/songs/other material, more translations, more tools, more spell/grammar checkers, etc.
  
• About a "clear situation", I guess we should have (*), among other things:
• Official language specs (thats okay, we got CLL 1.1)
• Official, clear, unique, and operational bug fixes workflow
  
• Implies to know where the "work in progress" branches are, have an view of what next version contains, etc.
• Organized proposals, with clear status, and with comparable annotations.
• That requires official and clear "fundamentals" too, I guess. Hence the previous try about it.
  
• Unique and centralized organization (ie. not 3 groups).
• Official and clear rules.
  
• etc.

(*) "We should have..." will be followed by proposals, of course.

@all: Any thoughts on my discussion? Any proposition?

Thanks.

la .sykyndyr.

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