Monthly Lojban Community Roundup — January 2017

This is the first of a series of monthly articles. The Monthly Lojban Community Roundup is a monthly roundup of community news and subjects that were discussed on IRC, the mailing list, reddit, Facebook and Twitter.

January 2017 started with, well, New Year’s Eve. Therefore, I will start out by showing some of the different ways people wished each other a good new year:

„za’a re no pa ze moi .i .a’o ro do ba lifri tu’a lo xamgu nanca“
„i ko’oi lo cnino nanca ba xamgu“
„di’ai cai doi lo jbopre .i doi la .lojban. ko renvi je cu banro“
„.i doi lo prenu .i gleki je cnino nanca“
„cnino detna’a di’ai doi ro do!“
„doi prenu .i ko gleki lo nu lo nanca cu cnino .i di’ai“

What is your favorite way of wishing someone a good time, a good new year, or a happy birthday? Let me know in the comments!

Right as the new year arrived, Lojban rap artist Djemynai sent out two tweets hinting at their plans for 2017:

We’re still waiting for further updates, but the prospect of a new album is exciting, and I think I’m not the only one who is curious about the new name.


Most people are probably familiar with sutysisku, which, next to vlasisku, is one of the most popular online dictionaries. This month, la gleki published a list which lists words according to the frequency with which they were looked up. Quite an interesting idea. As la gleki put it: “Why is this important? You can see what people were trying to find, which concepts are still absent in the dictionary and thus might need some work so that new users finally find highly needed words.”

You are probably wondering which words were the most common search queries according to the list. The most common gismu were cusku, tavla, klama, zukte and gasnu. Not surprising. The most commonly searched cmavo were ca, .i, ba, la, da, cu, ka, se, ta, ma, ci and lo. And if we can believe the list, the most frequently searched for English term was this thing:



I don’t know who entered toast this often, but at least it gave me something to write about. (inb4 toast memes?)


On Facebook, Superjoël la Pegazus asked: “What would be needed now to move Lojban forward ?” and received a lot of different responses. It was interesting to see what people thought could or should be done to help Lojban get bigger and better. Here’s a summary (for the whole discussion, follow the link above):

  • More and better learning material, beginner-friendly courses
  • Artistic work that promotes Lojban, as Star Trek did for Klingon
  • An updated CLL in a more modern style
  • Books, comics, videos (not only tutorials, but also vlogs), live streams.

All of these things would be nice to have. The more people help create content, the more people will get interested in the language, and the more people learn the language, the more people there will be who can help create even more content. Someone just has to begin! Let us be grateful for all the things we already have. Let us say thank you to anyone who wrote or contributed to learning materials, who made movies or songs, and also simply to anyone who has been studying the language. Keep at it, you are the future of Lojban.


Btw, if you are interested in knowing what words recently got added to jbovlaste, you can use its Recent Changes page. Here are all the words and definitions that were added in the last 30 days.


On Facebook, la gleki posted the following image:

The word for the knife-spoon is missing, for some reason. Anyway, I feel like this sort of thing is the perfect opportunity to create actual Lojban portmanteaus, rather than lujvo. The difference is that lujvo are limited to whatever rafsi their components have, and those rafsi aren’t always the most suitable candidates for making good-sounding portmanteaus. (In general, I am known to advocate portmanteau brivla and ignoring whether they are lujvo or zi’evla. A popular example that has caught on is pensku (based on pensi + cusku), which the traditional view would insist should be understood as penmycusku, because -pen- is a rafsi of penmi. More and more of such words are starting to show up. People are starting to release their fear of creative word invention.)

So let’s see what we can do here. The nice (but also difficult) thing with this freedom is that there are usually multiple possible ways a word can turn out, and we have to (or rather, are allowed to) choose, based on our subjective preferences. This makes word creation more personal, and more artistic. What are the options for the cutlery combinations?


dakfu + forca: dakforca, dakfo’a, forcfu, dakca, dorca, …

dakfu + smuci: daksmu, dakfuci, smukfu, smuda’u, …

forca + smuci: forcuci, forsmu, *smurca, …

dakfu + forca + smuci: daksmuca, fordaci, smuforfu, …

Which forms sound best to you? What can you come up with?


On IRC, adlai asked “Is there a bot translating naturalang to Lojban?”.

This question is probably of interest to a lot of people, so I’ll list the things that currently exist. We’ve had simple glossers for a long time, but it was only relatively recently (December 2015) that a statistical machine translator appeared: zmifanva, by Masato Hagiwara. Unlike a glosser, zmifanva tries “to give translation results that are as natural as possible”. It learns from Lojban-English sentence pairs on Tatoeba. The more sentence pairs are added to Tatoeba, and the higher their quality, the better zmifanva will become, so if you like this project, consider contributing to Lojban on Tatoeba.

As for the glossers, there are three that can be mentioned. The oldest and still the most sophisticated is the one that comes with jbofihe. There is an online version, which you can test. Its main drawback is the fact that it chokes on grammar that’s newer than 15 years, but the glossing isn’t bad. For example, it renders ro da zo’u mi prami da gi’e xebni ro nu da zukte lo’e xlali as:


A newer tool is this Lojban parser with an included glosser. It uses a much newer grammar (Ilmen‘s camxes-beta PEG) and has several output options. The Boxes tab (selectable after you hit Parse) is especially popular. It combines visual syntax with glossing and looks like this:


If you’re on IRC (#lojban, #ckule or #jbosnu), there is also a simple glosser implemented in mensi, the well-known allrounder bot. Type “gloss: “ and mensi will give a word by word translation:


And those are the tools we currently have.


Over on Discord, Jace ross asked: “hey new here anyone know any good learning vids”

If you speak Japanese, then you’re in luck, because ko lojbo iu by guskant is a full series of audio lessons. You can find the Youtube playlist here. The main page is here, which has links to pure audio files, as well as text versions of the lessons.

There is also this live stream by kribacr, which offers a slow introduction into the very basics of the language:

And last but not least, a well-made introduction to the Lojban bridi, by Dustin Lacewell:

If you are looking for listening-reading practice, search Youtube for Lojban corpus readings. Otherwise, just type in Lojban and watch everything there is, as I did.


On IRC, solpahi asked “Is there really no word for teapot?”

Wait, that’s me.

Is there really no word for teapot?!


la uakci published a short speech on reddit, in which they discuss the term “Official Lojban”. Starting from the point of view that there are many Lojbans, many dialects, all of which deserve to be called Lojban, la uakci argues that by calling one specific version “Official Lojban” all other versions get degraded and their users feel like they are using an inferior, less real, language. Instead, la uakci suggests we use the term “Base Lojban” or “Fundamental Lojban”.

I commented that there is a difference between Base Lojban and that which one could consider the common core that all Lojbans share. The latter is significantly smaller than Base Lojban, because many dialects come with replacements both of cmavo for other cmavo and with changes to syntax or morphology.

In response to this, la lalxu suggested the term «krasi lojbo» / “Original Lojban”, and la uakci agreed.

What do you think? Share your views in the comments.



( •_•)>⌐■-■



This month also saw a few new translations, which I will introduce to you below.

Make sure to check ’em all out.


The types of conversations Lojbanists have…



And lastly, as Twitter didn’t offer much in terms of subjects, I’ll just list my three favorite tweets:

Okay, the last one is technically from December, but it’s such a nice pun it had to get featured.

And that’s it for January. Now let’s see what February has in store for us. The more the people of Lojbanistan do, the more there will be for me to report. Take this as an incentive to contribute, to discuss, to learn, to ask questions. An incentive to help Lojban grow.

If you like this format, let me know!

Categories eng

9 thoughts on “Monthly Lojban Community Roundup — January 2017

  1. Wow! This is awesome! Thanks for writing this up.
    I think the guy who tweeted he thought lojban was terrible may have been trolling. It seems like the only things he tweets are just assertions which, while following a general pattern as far as political philosophy is concerned, seem to ill-substantiated to be serious invitations to debate. I don’t know though, but it feels like the whole account is a troll.


    1. Oh yes, it definitely is a troll account. I heard it tries to lure people into arguments. I just liked that particular tweet because it was especially inarticulate.
      Thank you for the nice comment.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s