| صفحات الموضوع: < [1 2 3 4 5 6 7] > | "Mini-contest" launched (on new "beta" contests platform) ناشر الموضوع: Henry Dotterer
|
|---|
| Three contest pairs going into a short qualification round | Oct 26, 2012 |
Hello all,
There was a better-than-expected response to calls for entries in some of the pairs Henry listed above. The minimum number of submissions in those pairs was far exceeded, and we've decided that a short qualification round is now needed to determine contest finalists.
... See more Hello all,
There was a better-than-expected response to calls for entries in some of the pairs Henry listed above. The minimum number of submissions in those pairs was far exceeded, and we've decided that a short qualification round is now needed to determine contest finalists.
The pairs are:
- English to Armenian
- English to Tagalog
- English to Thai
Anyone working or native in these languages is invited to participate in helping to select the finalists. This short qualification round will last until Monday, October 29th at 07:00 GMT.
Best regards,
Matt ▲ Collapse
| | | | | 58 pairs now in the finals | Oct 29, 2012 |
Thanks to those who responded to the call for ratings on short notice in those three pairs. A good number of ratings came in and it was possible to select finalists.
58 of the 60 pairs with the required minimum of three entries are now in the finals round. Two more pairs -- Spanish and Italian -- will go into finals voting shortly.
| | | | | some questions about the Finals | Oct 29, 2012 |
What means "Translations of this segment (34 total; 29 unique)? - What are "29 unique"?.
And how come that the lowes rating to be given for segments and entries alike, is 1? Why not 0? - Or even 0 down to -5? But wishing to give a bad translation a low rating and giving it a 1 such a piece or segment may get more than it deserves.
How is it that one sees now, in my combination, 7 "Finalists" on top, and all the other are "Non Finalist" posted below, but while for the "... See more What means "Translations of this segment (34 total; 29 unique)? - What are "29 unique"?.
And how come that the lowes rating to be given for segments and entries alike, is 1? Why not 0? - Or even 0 down to -5? But wishing to give a bad translation a low rating and giving it a 1 such a piece or segment may get more than it deserves.
How is it that one sees now, in my combination, 7 "Finalists" on top, and all the other are "Non Finalist" posted below, but while for the "Non Finalists" there is a - not very clear - breakdown of ratings got so far, e.g.
Rating type Overall Quality Accuracy
Segments 3.64 3.44 (146 ratings) 3.83 (157 ratings)
Entry 3.00 3.25 (4 ratings) 2.75 (4 ratings).
there is no such breakdown given for the "Finalists". -
Are the "Finalists" so good that they got 100% in all "Rating types"?
Where are the results that made the "Finalists" to "Finalists".
[Edited at 2012-10-30 13:05 GMT] ▲ Collapse
| | | | | learning to see the point in rating segments separately | Oct 30, 2012 |
Although it appeared to me very grotesque on first sight I now start to get the idea behind the rating option for segments separately. I see that this allows an in-depth, high precision rating of these segments individually. And by establishing immediately the average percentage after each new rating, either high or low, one certainly gets a very qualified vote.
And, after having rated the first segment completely, 34 entries in my combination, and started with the second segnment, ... See more Although it appeared to me very grotesque on first sight I now start to get the idea behind the rating option for segments separately. I see that this allows an in-depth, high precision rating of these segments individually. And by establishing immediately the average percentage after each new rating, either high or low, one certainly gets a very qualified vote.
And, after having rated the first segment completely, 34 entries in my combination, and started with the second segnment, I now see even some interest too in differentiating between an "accurate content rendering" and "poor writing". I saw indeed cases that had the content right but incorrect grammar or style level, inconsistencies etc.
Yet, to be precise. - It is possible for me to give a higher rating for "accuracy" and a lower for "quality of writing". But never the other way round, a higher rating for "quality of writing" than for "accuracy". Giving a 1 for "accuracy" means for me to give a 1 for "quality of writing" too. - If someone translates "the sky is blue" into "the "sky is green" or "pink" I would never give anything else than the lowest rate on either, "accuracy" and "quality of writing", no matter how precise the spelling of "green" and "pink" would be. If the translation is wrong, quality of writing does not matter any more. - But sure, "quality of writing" can make a difference between otherwise accurate or correct translaltions.
To take up the cooking analogy. - If, in a cooking contest, several plates taste equally nice, "quality of presentation" could become the match winner. But if a plate does not taste nice, or even horrible, "quality of presentation" does not even have to be taken into consideration any more.
If it is not too much work it may be useful that the system only allows better ratings for "quality or writing" if "accuracy of translation" is at least rated 2. - If "accruacy of translaltion" is rated 1 - as the lowest rate in the system - there should be no option for rating "quality of writing" other than 1 either.
[Edited at 2012-10-30 20:50 GMT] ▲ Collapse
| | |
|
|
|
| Finals Phase Bug? | Oct 30, 2012 |
The English to Japanese pair has moved to the finals round, but I can't seem to find any entries to vote on. I see the following:
Entries (8 total; 7 finalists)
Choose what you feel are the best three translations from those listed below. Note that for fairness, entries are shown in a different, random order to different viewers.
Th... See more The English to Japanese pair has moved to the finals round, but I can't seem to find any entries to vote on. I see the following:
Entries (8 total; 7 finalists)
Choose what you feel are the best three translations from those listed below. Note that for fairness, entries are shown in a different, random order to different viewers.
There is nothing below this. Is this some kind of bug, or am I looking in the wrong place? ▲ Collapse
| | | | | I'll check it out, Nathan | Oct 30, 2012 |
Nathan Takase wrote:
The English to Japanese pair has moved to the finals round, but I can't seem to find any entries to vote on. I see the following:
Entries (8 total; 7 finalists)
Choose what you feel are the best three translations from those listed below. Note that for fairness, entries are shown in a different, random order to different viewers.
There is nothing below this. Is this some kind of bug, or am I looking in the wrong place?
Thanks for reporting this. I'll look into it.
| | | | | Clarifications | Oct 30, 2012 |
Good questions, Roland.
Roland Nienerza wrote:
What means "Translations of this segment (34 total; 29 unique)? - What are "29 unique"?.
Some of the translations were identical. There are 29 distinct variations.
And how come that the lowes rating to be given for segments and entries alike, is 1? Why not 0? - Or even 0 down to -5? But wishing to give a bad translation a low rating and giving it a 1 such a piece or segment may get more than it deserves.
That's just how we did it. I understand the feeling (though on a relative scale 1 arguably serves the same purpose) and based on the usage patterns, suppose that there may be some people who are refraining from rating for that reason. So "tagging" will be restored, giving an option at least for flagging errors.
How is it that one sees now, in my combination, 7 "Finalists" on top, and all the other are "Non Finalist" posted below, but while for the "Non Finalists" there is a - not very clear - breakdown of ratings got so far, e.g.
Rating type Overall Quality Accuracy
Segments 3.64 3.44 (146 ratings) 3.83 (157 ratings)
Entry 3.00 3.25 (4 ratings) 2.75 (4 ratings).
there is no such breakdown given for the "Finalists". -
Are the "Finalists" so good that they got 100% in all "Rating types"?
Where are the results that made the "Finalists" to "Finalists".
So their ratings were, broadly speaking, better. The numbers are not shown for them so as to not bias finals round voting.
| | | | | But it is possible to rate quality of writing without rating accuracy of translation | Oct 30, 2012 |
Roland Nienerza wrote:
Although it appeared to me very grotesque on first sight I now start to get the idea behind the rating option for segments separately. I see that this allows an in-depth, high precision rating of these segments individually. And by establishing immediately the average percentage after each new rating, either high or low, one certainly gets a very qualified vote.
Yes, that is the idea.
And, after having rated the first segment completely, 34 entries in my combination, and started with the second segnment, I now see even some interest too in differentiating between an "accurate content rendering" and "poor writing". I saw indeed cases that had the content right but incorrect grammar or style level, inconsistencies etc.
Yet, to be precise. - It is possible for me to give a higher rating for "accuracy" and a lower for "quality of writing". But never the other way round, a higher rating for "quality of writing" than for "accuracy". Giving a 1 for "accuracy" means for me to give a 1 for "quality of writing" too. - If someone translates "the sky is blue" into "the "sky is green" or "pink" I would never give anything else than the lowest rate on either, "accuracy" and "quality of writing", no matter how precise the spelling of "green" and "pink" would be. If the translation is wrong, quality of writing does not matter any more. - But sure, "quality of writing" can make a difference between otherwise accurate or correct translaltions.
To take up the cooking analogy. - If, in a cooking contest, several plates taste equally nice, "quality of presentation" could become the match winner. But if a plate does not taste nice, or even horrible, "quality of presentation" does not even have to be taken into consideration any more.
If it is not too much work it may be useful that the system only allows better ratings for "quality or writing" if "accuracy of translation" is at least rated 2. - If "accruacy of translaltion" is rated 1 - as the lowest rate in the system - there should be no option for rating "quality of writing" other than 1 either.
I understand your argument and analogy. But consider that it is possible for -- and the contest platform now allows -- a person who is native in the target language to rate the quality of writing of an entry, without considering the accuracy of translation. (A native speaker of English can therefore rate the quality of writing of a Turkish to English translation, without understanding Turkish.) In principle, I believe this possibility should make the contests better.
| | |
|
|
|
| Bug preventing voting in Japanese should be fixed | Oct 30, 2012 |
Henry Dotterer wrote:
Nathan Takase wrote:
The English to Japanese pair has moved to the finals round, but I can't seem to find any entries to vote on. I see the following:
Entries (8 total; 7 finalists)
Choose what you feel are the best three translations from those listed below. Note that for fairness, entries are shown in a different, random order to different viewers.
There is nothing below this. Is this some kind of bug, or am I looking in the wrong place?
Thanks for reporting this. I'll look into it.
OK, should be fixed.
| | | |
I think formality of this contest should be dropped from "mini" to the "test" level.
Because it is a beta testing, but not a mini contest.
In Turkish there was a bug too.
The system prevented me to rate the entries several days before the end of the hybrid phase.
The same and similar bugs are happening everyday.
So I strictly advise a name change into "beta test contest" or "test contest".
It is obvious that we are testing but not contesti... See more I think formality of this contest should be dropped from "mini" to the "test" level.
Because it is a beta testing, but not a mini contest.
In Turkish there was a bug too.
The system prevented me to rate the entries several days before the end of the hybrid phase.
The same and similar bugs are happening everyday.
So I strictly advise a name change into "beta test contest" or "test contest".
It is obvious that we are testing but not contesting.
Announcing a "winner" in a test atmosphere will definitely be unfair for the past and future winners. ▲ Collapse
| | | |
Ali Bayraktar wrote:
I think formality of this contest should be dropped from "mini" to the "test" level.
Because it is a beta testing, but not a mini contest.
In Turkish there was a bug too.
The system prevented me to rate the entries several days before the end of the hybrid phase.
The same and similar bugs are happening everyday.
So I strictly advise a name change into "beta test contest" or "test contest".
It is obvious that we are testing but not contesting.
Announcing a "winner" in a test atmosphere will definitely be unfair for the past and future winners.
If there were any evidence that the contest were moving in a direction of selecting illegitimate winners, I would accept your suggestion. However, it seems clear that in the majority of pairs, at least, the contest is valid and the winners will be deserving. So my feeling would be that to not select winners would be unfair to current winners.
| | | |
Henry Dotterer wrote:
If there were any evidence that the contest were moving in a direction of selecting illegitimate winners, I would accept your suggestion.
It is testing.
The process of testing is already an evidence.
Simply saying the contest consists of two parts.
- Evaluation itself (takes place between the entries and the voters)
- Facilities (Software, web infrastructure that is used)
The words testing, beta means that you have new facilities and you are trying it for the first time.
When you use the same facility for the second time, ti wont be same with the first one.
There is an obvious difference here.
This difference should be reflected to the assigned titles and awards.
It must be mentioned that those winners used new facilities.
Facilities which is still in testing stage.
Facilities which still can not be trusted in 100%. (Beta has such meaning that it is not reliable in 100%)
I am not saying anything about entries. I believe that all contestants put a hard work in their entries. And respect that.
But this can not change the big picture.
The word Beta (Testing) can not be ignored here.
| | |
|
|
|
| I take your point | Oct 31, 2012 |
Ali Bayraktar wrote:
Henry Dotterer wrote:
If there were any evidence that the contest were moving in a direction of selecting illegitimate winners, I would accept your suggestion.
It is testing.
The process of testing is already an evidence.
...
The words testing, beta means that you have new facilities and you are trying it for the first time.
When you use the same facility for the second time, ti wont be same with the first one.
There is an obvious difference here.
... (Beta has such meaning that it is not reliable in 100%)
I can't disagree with your argument; everything you say here is true. It is also true, though, that we have improved the contests in each of our nine or ten revisions, and that no method will be 100% reliable. It really boils down to whether or not the ones that wind up being judged winners deserved to be so. In the overwhelming majority of pairs, rating and voting are "lining up", which is to say that the segment-based ratings, the entry-based ratings, and the finals round voting are converging, so that the leading entries have been judged to be leaders by different people using different methods.
That said, there are a few pairs out of the sixty where this is not yet clearly the case. I'm watching those.
I am not saying anything about entries. I believe that all contestants put a hard work in their entries. And respect that.
Right. And ratings would suggest that some are clearly very good.
| | | | | Spanish and Italian are now open for voting | Oct 31, 2012 |
The last two pairs with at least three entries have been opened for voting -- Spanish and Italian.
That makes 60 pairs with active competitions.
| | | |
Henry Dotterer wrote:
Henry Dotterer wrote:
Nathan Takase wrote:
The English to Japanese pair has moved to the finals round, but I can't seem to find any entries to vote on. I see the following:
Entries (8 total; 7 finalists)
Choose what you feel are the best three translations from those listed below. Note that for fairness, entries are shown in a different, random order to different viewers.
There is nothing below this. Is this some kind of bug, or am I looking in the wrong place?
Thanks for reporting this. I'll look into it.
OK, should be fixed.
Thank you - it is indeed now working properly.
| | | | | صفحات الموضوع: < [1 2 3 4 5 6 7] > | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » "Mini-contest" launched (on new "beta" contests platform) | CafeTran Espresso | You've never met a CAT tool this clever!
Translate faster & easier, using a sophisticated CAT tool built by a translator / developer.
Accept jobs from clients who use Trados, MemoQ, Wordfast & major CAT tools.
Download and start using CafeTran Espresso -- for free
Buy now! » |
| | 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 » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
