Pages in topic:   < [1 2 3] >
Contests: Congratulations to winners in "A translator's life"!
Thread poster: Ana Moirano
Zea_Mays
Zea_Mays  Identity Verified
Italië
Local time: 04:13
uit Engels in Duits
+ ...
@Barbara Oct 16, 2023

I agree with you about the quality of the profile.
The bot entry presentation was done by Jared and he also said that he filled in the MT output in the contest entries.
I don't know if he also filled the bot's profile.
Honestly, I don't see a big problem with these gimmicks and their subpar profiles (since they are gimmicks, they can have bad profiles, why should they have good ones?). Having Chatty in KudoZ is quite worse.
As said, if MT is better than human translation
... See more
I agree with you about the quality of the profile.
The bot entry presentation was done by Jared and he also said that he filled in the MT output in the contest entries.
I don't know if he also filled the bot's profile.
Honestly, I don't see a big problem with these gimmicks and their subpar profiles (since they are gimmicks, they can have bad profiles, why should they have good ones?). Having Chatty in KudoZ is quite worse.
As said, if MT is better than human translation in some cases, maybe that says something about the quality of translation some of us provide, doesn't it? In all of Proz's contests you'll see entries where you wonder how on earth the entrant can call themselves a "translator".
Collapse


Josephine Cassar
Lieven Malaise
Barbara Carrara
P.L.F. Persio
 
Barbara Carrara
Barbara Carrara  Identity Verified
Italië
Local time: 04:13
Member (2008)
uit Engels in Italiaans
+ ...
Zea Oct 16, 2023

Zea_Mays wrote:
The bot entry presentation was done by Jared and he also said that he filled in the MT output in the contest entries.


How do you know that? Did Jared explicitly say so? Where?



Honestly, I don't see a big problem with these gimmicks and their subpar profiles (since they are gimmicks, they can have bad profiles, why should they have good ones?).


Don't you find 'its' chosen wording infuriating, and 'its' sarcasm uncalled for?


Having Chatty in KudoZ is quite worse.


Well THAT is definitely infuriating. However, I cannot find any KudoZ page for the 'GoTrans' bot, but only its many contest translations out of and into many language pairs, included in its hundred-pair rich list...

I'd very much like to read reviews from colleagues working in any of those pairs... And from Jared.


 
Zea_Mays
Zea_Mays  Identity Verified
Italië
Local time: 04:13
uit Engels in Duits
+ ...
bot Oct 16, 2023

Barbara Carrara wrote:

Zea_Mays wrote:
The bot entry presentation was done by Jared and he also said that he filled in the MT output in the contest entries.


How do you know that? Did Jared explicitly say so? Where?

During the 2023 Translation Day event (the recording is available, I think for plus subscribers, you'll find the link in one of my previous posts).
The bot has no KudoZ history: https://www.proz.com/profile/1127212?show_mode=kudoz
(just replace your profile number in your own KudoZ history with the number of another user, and you'll get there)

[Bearbeitet am 2023-10-16 11:07 GMT]


Barbara Carrara
 
Christel Zipfel
Christel Zipfel  Identity Verified
Local time: 04:13
uit Italiaans in Duits
+ ...
@ Zea_Mays: So, if I understand correctly.... Oct 16, 2023

Zea_Mays wrote:


Likes/dislikes do not count for ranking.
IMO, likes/dislikes can be quite misleading as often they are very subjective.
At least a third, "orange" tag would be required for preferential notes here.



Let's take as an example Italian-German where there were 3 entries, including yours. The other two entries have got 21 dislikes each one and yours none. I agree that likes are subjective, but as far as mistranslations or grammar/spelling mistakes (= dislikes) are concerned, there is little or no room at all for subjectivity. So, if one or both other entries had got more points than yours, it would no longer have appeared as the best one, would it? Here, I am really flabbergasted! I mean, voting is very time consuming, so why would anyone vote and provide comments and possibly links and spend all this time, if it's true that likes/dislikes do not count for ranking? I can tell after only a few sentences, and without checking word by word, whether I feel a translation is good/better than others and whether I "like" or "dislike" it, which would be indeed very subjective, though! And it's just a matter of minutes!
Thank you for having enlightened me.


Zea_Mays
 
Lingua 5B
Lingua 5B  Identity Verified
Bosnië en Herzegowina
Local time: 04:13
Member (2009)
uit Engels in Kroaties
+ ...
Everything is subjective Oct 16, 2023

In case of a contestant also being a voter, everything is subjective even more.

Of course likes/dislikes don't count, especially dislikes, if they come from other contestants. It's a pure conflict of interest, no rocket science.

IMO, contestants should be banned from voting.

I once skimmed the situation in my language pair, and I saw a heap of dislikes and very few likes anywhere, mostly none of course. It was pretty clear who the voters were, their compet
... See more
In case of a contestant also being a voter, everything is subjective even more.

Of course likes/dislikes don't count, especially dislikes, if they come from other contestants. It's a pure conflict of interest, no rocket science.

IMO, contestants should be banned from voting.

I once skimmed the situation in my language pair, and I saw a heap of dislikes and very few likes anywhere, mostly none of course. It was pretty clear who the voters were, their competitors. I have never participated in a contest, but it's easy to spot the culture and what's going on there if you just briefly visit the contest page.

There should be a standardized and clear voting criteria sheet against which this should be rated. This sheet should not be shown to participants, as they will find a way to abuse anything/everything and game the system.





[Edited at 2023-10-16 13:09 GMT]
Collapse


Zea_Mays
Elena Feriani
Maria Teresa Borges de Almeida
 
Zea_Mays
Zea_Mays  Identity Verified
Italië
Local time: 04:13
uit Engels in Duits
+ ...
likes/dislikes do not matter Oct 16, 2023

Christel Zipfel wrote:

Let's take as an example Italian-German where there were 3 entries, including yours. The other two entries have got 21 dislikes each one and yours none. I agree that likes are subjective, but as far as mistranslations or grammar/spelling mistakes (= dislikes) are concerned, there is little or no room at all for subjectivity. So, if one or both other entries had got more points than yours, it would no longer have appeared as the best one, would it?

Yes. If one of them had managed to find let's say 5 fellow translators to vote for their entry with highest vote (mine was voted by 4), it would appear as the best one.


 
Zea_Mays
Zea_Mays  Identity Verified
Italië
Local time: 04:13
uit Engels in Duits
+ ...
bot entries Oct 16, 2023

Barbara Carrara wrote:
(I can only see) its many contest translations out of and into many language pairs, included in its hundred-pair rich list...

Where do you see them?


 
Barbara Carrara
Barbara Carrara  Identity Verified
Italië
Local time: 04:13
Member (2008)
uit Engels in Italiaans
+ ...
GoTrans translation contest entries Oct 16, 2023

Zea_Mays wrote:

Where do you see them?


Whilst searching online for the missing bot KudoZ history, I stumbled upon a list of old ProZ translation contest entries uploaded (by whom, on the bot behalf?) over the years.

If you google "proz GoTrans translation contest entries", 18 of them come up in the first couple of pages, from different years and in a number of language pairs.

For the record, I did not open any of them.


Maria Teresa Borges de Almeida
expressisverbis
 
Zea_Mays
Zea_Mays  Identity Verified
Italië
Local time: 04:13
uit Engels in Duits
+ ...
@Barbara Oct 16, 2023

I see. These are the single entries, each of which gets its own page.
I also found a blog article about the bot: https://go.proz.com/blog/human-response-to-a-changing-industry-part-1
Seems to be from before the last contest.


Barbara Carrara
Maria Teresa Borges de Almeida
expressisverbis
 
Christel Zipfel
Christel Zipfel  Identity Verified
Local time: 04:13
uit Italiaans in Duits
+ ...
Disagree Oct 16, 2023

Lingua 5B wrote:

In case of a contestant also being a voter, everything is subjective even more.

Of course likes/dislikes don't count, especially dislikes, if they come from other contestants. It's a pure conflict of interest, no rocket science.


Dislikes, whether they come from a contestant or not, are incontestable as long as they refer to grammar/syntax/spelling errors, which are always documentable, and hence not subjective at all. Therefore they can absolutey not be ignored.


 
Lingua 5B
Lingua 5B  Identity Verified
Bosnië en Herzegowina
Local time: 04:13
Member (2009)
uit Engels in Kroaties
+ ...
Disagree Oct 16, 2023

Christel Zipfel wrote:

Lingua 5B wrote:

In case of a contestant also being a voter, everything is subjective even more.

Of course likes/dislikes don't count, especially dislikes, if they come from other contestants. It's a pure conflict of interest, no rocket science.


Dislikes, whether they come from a contestant or not, are incontestable as long as they refer to grammar/syntax/spelling errors, which are always documentable, and hence not subjective at all. Therefore they can absolutey not be ignored.


Can you find me another contest in anything in the world where contestant is also a voter? That's exactly the problem. The voting criteria is chaotic and lacks structure, where a spelling error has the same weight as a semantic error, for instance.

[Edited at 2023-10-16 20:27 GMT]


Josephine Cassar
Maria Teresa Borges de Almeida
P.L.F. Persio
 
Maria Teresa Borges de Almeida
Maria Teresa Borges de Almeida  Identity Verified
Portugal
Local time: 03:13
Member (2007)
uit Engels in Portugees
+ ...
GoTrans Oct 17, 2023

So, let's summarise the situation: the translators who are submitting entries to the translation contest each year are feeding GoTrans by giving it the chance to get better and better. Machiavellian or what?

expressisverbis
Barbara Carrara
P.L.F. Persio
 
Zea_Mays
Zea_Mays  Identity Verified
Italië
Local time: 04:13
uit Engels in Duits
+ ...
Not really Oct 17, 2023

Maria Teresa Borges de Almeida wrote:

So, let's summarise the situation: the translators who are submitting entries to the translation contest each year are feeding GoTrans by giving it the chance to get better and better. Machiavellian or what?


GoTrans "delivered" unedited DeepL output until the forelast contest. In the latest contest the unedited output was taken from ChatGPT.
So I think these AI features do not profit from the contest. GoTrans only shows the improvement of MT over the years.


 
Christel Zipfel
Christel Zipfel  Identity Verified
Local time: 04:13
uit Italiaans in Duits
+ ...
..,... Oct 17, 2023

Lingua 5B wrote:

Can you find me another contest in anything in the world where contestant is also a voter? That's exactly the problem. The voting criteria is chaotic and lacks structure, where a spelling error has the same weight as a semantic error, for instance.


I didn't say it is ok that contestants can also vote, but I disagree with your affirmation that behind their dislikes, as far as "documentable" mistakes are concerned, there is necessarily some personal interest, as those dislikes cannot be considered "subjective".


 
Zea_Mays
Zea_Mays  Identity Verified
Italië
Local time: 04:13
uit Engels in Duits
+ ...
Why agrees & disagrees cannot be counted in Oct 19, 2023

Christel Zipfel wrote:

I didn't say it is ok that contestants can also vote, but I disagree with your affirmation that behind their dislikes, as far as "documentable" mistakes are concerned, there is necessarily some personal interest, as those dislikes cannot be considered "subjective".


Regarding errors, disagreements are not subjective, I agree.
I have noticed that often multiple people tag the same item multiple times,
instead of agreeing with who had already tagged the same item.
This makes an entry look full of red tags and therefore really wrong to the eye, while this is not always the case.
This can be misleading.

I have also counted several cases where disagrees are really subjective or even wrong,
not only on my own translations.
It wouldn't be fair counting these in without vetting. And all disagrees to a disagree would have to be subtracted.
Agrees can also be very subjective, and people are much more stingy with them. Also here, disagrees would have to be subtracted.

As said, I'd suggest to have at least a third tag: orange for preferential comments.
And I would be in favour of the system recognising that a term has already been tagged,
and giving you just the option to agree or disagree and comment.

Another issue may be the use of mobile devices.
I already find it difficult and cumbersome to read and rate all entries on a desktop, but how will it be when done on a smartphone?
The more entries there are, the harder this is, and the more a fair chance may be lacking.



[Bearbeitet am 2023-10-19 09:51 GMT]


Lingua 5B
 
Pages in topic:   < [1 2 3] >


To report site rules violations or get help, contact a site moderator:

Moderator(s) of this forum
Lucia Leszinsky[Call to this topic]

You can also contact site staff by submitting a support request »

Contests: Congratulations to winners in "A translator's life"!






Trados Business Manager Lite
Create customer quotes and invoices from within Trados Studio

Trados Business Manager Lite helps to simplify and speed up some of the daily tasks, such as invoicing and reporting, associated with running your freelance translation business.

More info »
Anycount & Translation Office 3000
Translation Office 3000

Translation Office 3000 is an advanced accounting tool for freelance translators and small agencies. TO3000 easily and seamlessly integrates with the business life of professional freelance translators.

More info »