Super Challenge: questions and discussion

[german2k01]

Just one more question. Regarding tweeting books and films. Should It be done at the end of each month or as soon as you are done watching a movie or reading a book you send a tweet about it?

[rdearman]

Just one more question. Regarding tweeting books and films. Should It be done at the end of each month or as soon as you are done watching a movie or reading a book you send a tweet about it?

You can tweet whenever you want, just don't do it before the 1st of May.

[IronMike]

hahaha! The bot read #Slavic as Slovenian. So I had to #giveup (Thanks Judith). I'll wait a few days to see if anyone takes Rick on for developing language family codes. If not, I'll just register for Old Church Slavonic or smthg.

[rdearman]

hahaha! The bot read #Slavic as Slovenian. So I had to #giveup (Thanks Judith). I'll wait a few days to see if anyone takes Rick on for developing language family codes. If not, I'll just register for Old Church Slavonic or smthg.

It has been a while but I think I did some kind of workaround last time and changed the 3 letter codes to numbers. I will look at at it later.

[Sprachprofi]

Thanks for setting this up!! Looking forward to challenging you all! I shall also promote this in a couple communities I'm part of.

Registration bug: I thought I remembered that #half challenge was a thing, but the bot interpreted it as Hausa, even though I had specified #Croatian earlier. #half is not a language code though. I'm wondering if the bot just takes the first two letters of any hashtag in order to guess the language. That would explain why three-letter codes are not accepted (or rather, are treated as if they were only two letters). In any case, it would be good to fix the underlying regular expression so #.. is matched against the database of two-letter codes, #... is matched against the database of three-letter codes and the rest is assumed to be either a full language name or irrelevant extra hashtags.

[zenmonkey]

OK, twitter unlocked the language challenge account and the website works. It is hosted somewhere else to keep traffic down on this site. It has a funky domain name, but it was just a domain I own for no good reason.

https://anki-tv.co.uk/

Registration instructions are on the site.

You can register at any time.
However !!!!

DO NOT TWEET ANY SCORES BEFORE THE FIRST OF MAY, BECAUSE I WILL BE CLEARING ALL THE ENTRIES AT MIDNIGHT ON THE 30TH AND YOU WILL LOSE YOUR SCORES.

I'm fairly certain that we never got the 3 letter language family thing working, although they are listed as options. e.g. egx = Family-Egyptian. Don't use them. If someone wants to fix this and you know PHP feel free to fork the code on github and fix it.

https://github.com/language-learners/superchallengebot

Quoting because it deserves it.
Thanks for all the work, R!

Now that you’ve done that … well, can’t let it go to waste.
I think the challenge concept can be useful, I don’t need the rules and I’m not a fan of the twitter bots but I’ll likely sign up. Thanks.

[rdearman]

Thanks for setting this up!! Looking forward to challenging you all! I shall also promote this in a couple communities I'm part of.

Registration bug: I thought I remembered that #half challenge was a thing, but the bot interpreted it as Hausa, even though I had specified #Croatian earlier. #half is not a language code though. I'm wondering if the bot just takes the first two letters of any hashtag in order to guess the language. That would explain why three-letter codes are not accepted (or rather, are treated as if they were only two letters). In any case, it would be good to fix the underlying regular expression so #.. is matched against the database of two-letter codes, #... is matched against the database of three-letter codes and the rest is assumed to be either a full language name or irrelevant extra hashtags.

That is exactly the problem. I did look into it and forked one a couple of years ago to fix it, but surrealix when he wrote it uses it everywhere, and it is a bit like digging through spaghetti to pull out the strand you want. Also, he created a lot of user-defined functions in the MySQL DB and the regular expression stuff (if I remember correctly) was tied up with these. I've done PL/SQL in Oracle and PGPL/SQL in PostgreSQL, but never bothered to look into the MySQL stuff. If I get a chance, I'll fork it and fix it, otherwise I'll probably just change the families to numeric codes, which is how I worked around it last time.

Three letter families are a bit of an edge-case since the only people I ever heard who want it were IronMike and Serpent.

[german2k01]

Need to clarify one more thing. For example, I finish reading a long novel (Mr.Mercedes)by Stephen King, 500 pages." Till the end.
So which command should I use? The first one or the second one. In the case of the first one, how twitter bot will know if this particular book/novel is 500 pages long and will automatically be split into the appropriate number against my score. This part is kind of vague for me as I am not mentioning pages in the first command. Can you show me how would you write it as a tweet? Thanks

@langchallenge I just read a #book." (The First One)

"@langchallenge I just finished reading a #sv #book with 60 pages. It was called "The Invisible Book" and I kept losing it."

(The Second One)

Reading : #book/#read, [#swedish/#french/#it/#ru/etc], [123 page/s], ["book title"]
A book automatically counts as 50 pages. If you only read part of a book, simply write how many pages you read, followed by the word 'page(s)'.
If you read a really long book, it will automatically be split into the appropriate number against your score.
If you want to say which book you've been reading, encase the title in quotation marks.
If you're studying multiple languages, you need to specify which one using the language code.
Examples: "@langchallenge I just read a #book."
"@langchallenge I just finished reading a #sv #book with 60 pages. It was called "The Invisible Book" and I kept losing it."

[IronMike]

Three letter families are a bit of an edge-case since the only people I ever heard who want it were IronMike and Serpent.

Anna Popkova figured one out, at least according to Twitter. Two hours ago she registered for family Semitic with #sem.

I'm going to try a few things. I can always un-register...

[Mathias]

Need to clarify one more thing. For example, I finish reading a long novel (Mr.Mercedes)by Stephen King, 500 pages."

Well, this is my first rodeo too, but for what I got:

Using "@langchallenge I just read a #book." will add 50 pages.
The bot can't know what or how much you read besides that here we're calling 50 pages a book, so this is what the bot knows.
If you want more than that, more information has to be provided following the model

Reading : #book/#read, [#swedish/#french/#it/#ru/etc], [123 page/s], ["book title"]

Something like: @langchallenge #book #german 500 pages "Mr Mercedes"

The second example just make your tweet look nicer, more like a proper sentence than a bunch of hashtags.