New feature – language mismatch detection ?
Some good news to report !
I am happy to see these, and to know that our Reddit community had a role in bringing these to life.
About two months ago we had some discussions about people receiving 'letters' (more like messages, indeed) in languages they have no knowledge of, and which are not listed in their profiles at all.
Machine translation ?
The banner image above is an example, one of various seen in topics in our sub-Reddit. If we processed it using a machine translation, the original could more or less be understood :
Asian languages are much more difficult to translate, compared to European and Western ones. The results above, from a DeepL translation, are not bad.
Would this be acceptable?
After some discussions in a couple of topics, I made a poll, asking our users what their personal opinion was about this. Was it ok, for a sender to not put any effort into matching a recipient's listed Language options?
The poll was open for a full week, and a total of 212 unique users voted. Snapshot of the results:
I was very pleased with the topic, the discussions and comments, user participation.
The Poll Results are clear!
Results are clear – 42.5 % of respondents said it was NOT OK, and they would Decline or even Report the letter as Spam.
Only 16 % of respondents considered it acceptable, and ok for the recipient to have to handle a Translation on their end.
My own view is that this is NOT acceptable at all, and it should be mandatory for a sender to write in a language which is listed on a Recipient's profile.
I posted a Twitter thread about this, and included the @SlowlyApp account.
New Features in the works?
There are signs of new work being placed into this by the SLOWLY development team. I will not share screenshots, but I have seen new dialogue screens on the Translation platform – which indicate the code is more or less done, and being rolled out in a Beta version sometime soon.
The new feature includes a Language Detection filter - SLOWLY itself will try to detect what language is being used in a new letter.
And once this is done, it compares that language with BOTH the sender's and the Recipient's Language options (as listed in their user profiles).
At this point a few different scenarios are possible :
- If a Sender is using a language not included in their own profile, a dialogue will appear – suggesting perhaps revising their profile to add that new lingo.
- If a Sender is using a language not included in the Recipient's profile, a dialogue will appear – suggesting it's unlikely the recipient would accept and reply to it.
- If a Sender is using a language which is included in a Recipient's profile, but at a 'Beginner' level, a dialogue will pop up and alert the sender, mention this and ask them to reconsider the choice of language for a better match.
Another nice new addition is that from within an Automatch search, a sender will now be able to visit the possible recipient's user profile. Which can be helpful, so they can have an idea on who would be receiving it, and if they think it appropriate to send that letter – or not.
Beta testing ? Spotted these features ?
For those of you who are on the Beta testing programme, please comment and report when you see these features appearing on a Beta release - along with the release version number.
And a big thank you to everyone who participated in the poll and contributed comments.
Famous Last Words
Special Thank You to the folks behind this wonderful app — all the hard working people at Slowly Communications.
Letters we share, with pen pals all over the World, via the magic of the Slowly App. A modern day take in the traditional penpal experience.
Come and join us, if you are not yet using Slowly.
Follow Blog via your Fediverse/Mastodon account :
@friends-near-and-afar-letters-we-share@wordsmith.social
This page created entirely in MarkDown language. Thank you for reading, feel free to comment, via a direct message to my Twitter account – or the Reddit one.
You can also post a public comment in the Reddit thread for it here.
