Pages in topic:   [1 2 3 4] >
Don't allow asker to select an answer which has not received any agrees or any comments at all
Thread poster: Bernhard Sulzer
Bernhard Sulzer
Bernhard Sulzer  Identity Verified
United States
Local time: 20:26
English to German
+ ...
Jan 4, 2015

Hopefully, this will discourage the simple copying (= very similar answer) of other people's answers and prevent that inexperienced askers choose that answer before any other colleagues have commented on it.
This would also work for the case where there is only one answer.
I wouldn't want to do all the work (explanation, links, etc.), then someone else comes along, enters something very similar and gets the points. Granted, inexperienced askers are more likely to simply choose an ans
... See more
Hopefully, this will discourage the simple copying (= very similar answer) of other people's answers and prevent that inexperienced askers choose that answer before any other colleagues have commented on it.
This would also work for the case where there is only one answer.
I wouldn't want to do all the work (explanation, links, etc.), then someone else comes along, enters something very similar and gets the points. Granted, inexperienced askers are more likely to simply choose an answer without considering the person who did all the legwork. But that's no excuse.
Your thoughts




[Edited at 2015-01-04 22:15 GMT]

Add-on: by "don't allow asker to select" I mean don't give them the option to select and award points. And if there's no answer or one or more answers but with no agrees, I suggest instituting a time limit after which the question will close automatically and no points will be awarded.

[Edited at 2015-01-05 00:45 GMT]
Collapse


 
Fiona Grace Peterson
Fiona Grace Peterson  Identity Verified
Italy
Local time: 01:26
Italian to English
Disagree Jan 4, 2015

Perhaps this is a problem in your language pair?

Anyone indulging in that kind of behaviour in my language pair gets weeded out pretty quickly.

Of course there are always going to be people who abuse a system, any system. And KudoZ is no exception. And it is frustrating to see answers with few or no references or helpful information chosen over those of other askers who have put time and effort into answering.

Let's not forget, however, that KudoZ is not ab
... See more
Perhaps this is a problem in your language pair?

Anyone indulging in that kind of behaviour in my language pair gets weeded out pretty quickly.

Of course there are always going to be people who abuse a system, any system. And KudoZ is no exception. And it is frustrating to see answers with few or no references or helpful information chosen over those of other askers who have put time and effort into answering.

Let's not forget, however, that KudoZ is not about points-grabbing; it's about helping others and showing your expertise in a given field (yes, really!!). And no-one can take that away, however unscrupulous or inexeperienced.

My advice would be to ignore it and let the guilty parties chase the points. There is more to life than KudoZ.
Collapse


 
Bernhard Sulzer
Bernhard Sulzer  Identity Verified
United States
Local time: 20:26
English to German
+ ...
TOPIC STARTER
More to life, yes! But ... Jan 4, 2015

Fiona Peterson wrote:

There is more to life than KudoZ.


You said it.

But ... those points sure help for being listed farther up in the translators directory.
In any case, I don't need the points anymore, I do it for fun, but I am going to think twice about whom I help out.
If one's efforts are completely ignored, it's certainly no fun. Not saying it happens all the time, but the system can be improved I believe, as long as ProZ.com is willing.

Thanks for your comments.


 
Lincoln Hui
Lincoln Hui  Identity Verified
Hong Kong
Local time: 08:26
Member
Chinese to English
+ ...
Thoroughly absent in reality or intelligence Jan 5, 2015

Chinese to English, first page. Of 23 closed questions with only one answer, 15 have no agrees. I have no reason to believe that any of the answers are incorrect.

Obviously I don't go into questions that are out of my realm. On the other hand, when the question is a simple one and the only answer provided is obviously correct, I don't bother leaving a comment either. This is generally the case with both my main language pairs, as it should be, and I will not stand for things to be
... See more
Chinese to English, first page. Of 23 closed questions with only one answer, 15 have no agrees. I have no reason to believe that any of the answers are incorrect.

Obviously I don't go into questions that are out of my realm. On the other hand, when the question is a simple one and the only answer provided is obviously correct, I don't bother leaving a comment either. This is generally the case with both my main language pairs, as it should be, and I will not stand for things to be changed.

The more specialized a field is, the less likely that it will or should get agrees. Don't allow answers in specialized fields to get points - fantastic idea.

We got termites in our house. Let's move the termites to a safe place, burn the house down, and put the termites in our new house. I can't think of a better way to screw the majority of Kudoz answerers and ensure that no one will ever answer questions on Kudoz again, especially in minor language pairs and specialized fields.
Collapse


 
Bernhard Sulzer
Bernhard Sulzer  Identity Verified
United States
Local time: 20:26
English to German
+ ...
TOPIC STARTER
I want confirmation Jan 5, 2015

Lincoln Hui wrote:

Thoroughly absent in reality or intelligence

????

Lincoln Hui wrote:
Chinese to English, first page. Of 23 closed questions with only one answer, 15 have no agrees. I have no reason to believe that any of the answers are incorrect.


There is no reason to believe they are correct either.

Lincoln Hui wrote:
Obviously I don't go into questions that are out of my realm. On the other hand, when the question is a simple one and the only answer provided is obviously correct, I don't bother leaving a comment either. This is generally the case with both my main language pairs, as it should be, and I will not stand for things to be changed.


You can certainly do what you want, but answers with neither agree nor comment, selected by the asker, awarded points and entered into the glossary are not what I would like to see in KudoZ. Goes for simple or more difficult questions.


Lincoln Hui wrote:
The more specialized a field is, the less likely that it will or should get agrees. Don't allow answers in specialized fields to get points - fantastic idea.


I hold that any answer should get agrees before it is selected and entered into the glossary. There ought to be assessments of answers by other colleagues to ensure that an answer is indeed correct. It doesn't help if someone thinks the answer is correct but doesn't comment.

Lincoln Hui wrote:
We got termites in our house. Let's move the termites to a safe place, burn the house down, and put the termites in our new house. I can't think of a better way to screw the majority of Kudoz answerers and ensure that no one will ever answer questions on Kudoz again, especially in minor language pairs and specialized fields.


I would want confirmation and support before I expect my answer to be chosen. And as an asker, I would want confirmation as well.
And I don't see your analogy as being applicable.

[Edited at 2015-01-05 05:48 GMT]


 
Andrea Halbritter
Andrea Halbritter  Identity Verified
France
Local time: 01:26
French to German
+ ...
Agree with Lincoln Hui Jan 5, 2015

For the pair French to German it is often the same as for Chinese to English. So I agree with Lincoln Hui. There often are no agrees at all. Even peers that agreed in the discussion do not agree directly on the answers.

On the other hand one or two peer agrees do not assure the answer is correct neither. 3 or 4 may do but it also depends on if people who agree actually work in the field, if they are natives or not etc. Although there are a lot of really excellent answers it already
... See more
For the pair French to German it is often the same as for Chinese to English. So I agree with Lincoln Hui. There often are no agrees at all. Even peers that agreed in the discussion do not agree directly on the answers.

On the other hand one or two peer agrees do not assure the answer is correct neither. 3 or 4 may do but it also depends on if people who agree actually work in the field, if they are natives or not etc. Although there are a lot of really excellent answers it already happened to me as well to see answers with three peer agrees which turnt out to be false.

So your proposition makes no sense to me and it is not possible to regulate everything. In the glossaries there will always be terms which are not correct. It's up to every single person who as a look at them to reverify.

[Modifié le 2015-01-05 06:49 GMT]
Collapse


 
Balasubramaniam L.
Balasubramaniam L.  Identity Verified
India
Local time: 05:56
Member (2006)
English to Hindi
+ ...
SITE LOCALIZER
This would not be a wise or useful change Jan 5, 2015

In my language pair (English to Hindi) people evaluate answers not by the linguistic merit in the answers, but by who is the asker and who are the other answerers. The asker himself, often takes into account various non-linguistic criteria to pick an answer. Some of these are:

1. Who is the answerer? If he happens to be a major outsourcer from whom he can expect to get work, he will pick his answer irrespective of any number of better and more correct answers being present.
... See more
In my language pair (English to Hindi) people evaluate answers not by the linguistic merit in the answers, but by who is the asker and who are the other answerers. The asker himself, often takes into account various non-linguistic criteria to pick an answer. Some of these are:

1. Who is the answerer? If he happens to be a major outsourcer from whom he can expect to get work, he will pick his answer irrespective of any number of better and more correct answers being present.

2. How many kudoz points do the answerers have? If the answerer has about as many kudoz points as the asker, and by selecting his answer, this answerer will to move above the asker in the kudoz totem pole, the asker will never ever select the answerer's answer as this would involve jeopardizing his own translation business. The asker will search for an answerer who has much less kduoz points than him so that he is no threat; or he will choose an answerer who is so high in the kudoz totem pole that it really doesn't matter if he earns a few more points, as for all practical purposes, the asker cannot aspire to beat him in kudoz points in the near future.

The agrees to an answer are also no reflection of the linguistic correctness or superiority of the answer. The same criteria as mentioned above are used to put agrees to an answer.

This is because people are afraid that if they put an agree to a competitor's answer, that answer might get selected, either by the asker or by the robot on the basis of maximum agrees an answer has. This might make the answerer move up the kudoz pole and above the person who has agreed to his answer!

Translators in the English-Hindi kudoz forum have found many other ingenious ways to run down or beat the system. One useful loop-hole in the kudoz rules that is put to devastating use is to mark a pro question into a non-pro question if a competitor's answer looks likely to be selected as the correct answer. This trick is most commonly used by those who are already up in the kudoz pole and want to push off other competitors who are snapping at their heels. If the latter translators answer any question correctly, these questions are immediately flagged as non-pro by the kudoz leaders. The points earned by non-pro questions are not used for translator ranking.

In fact, the non-proz questions are more fairly and intelligently agreed to and evaluated, as they are of a non-threatening nature.

So there is a lot of politics being played in the kudoz forums (at least in the English to Hindi one) and taking into account the agrees to answers will in no way improve the quality of the kudoz glossary.

The only way to improve kudoz is to stop using kudoz points for ranking translators on this site.

[Edited at 2015-01-06 03:24 GMT]
Collapse


 
Heinrich Pesch
Heinrich Pesch  Identity Verified
Finland
Local time: 02:26
Member (2003)
Finnish to German
+ ...
Will not work in rare language pairs Jan 5, 2015

When I put a question into FI-DE kudoz, I pretty much can be sure only one person will answer, even if I wait for ages. I usually close my question when the system reminds me, after one week.
Perhaps it is better to restrict closing questions after a waiting period of 24 hours or so.


 
Thomas Pfann
Thomas Pfann  Identity Verified
United Kingdom
Local time: 00:26
Member (2006)
English to German
+ ...
Counter-productive Jan 5, 2015

Further to the points already raised by others, I can think of several common scenarios in which it would be counter-productive to not allow the asker to select an answer without agrees:

- Questions which seem obvious at first but have a finer twist to them which is not realized until late. People might be on the wrong track first and incorrect answers might get agrees until someone has a lightbulb moment and posts the correct answer. This answer might not get any more agrees becaus
... See more
Further to the points already raised by others, I can think of several common scenarios in which it would be counter-productive to not allow the asker to select an answer without agrees:

- Questions which seem obvious at first but have a finer twist to them which is not realized until late. People might be on the wrong track first and incorrect answers might get agrees until someone has a lightbulb moment and posts the correct answer. This answer might not get any more agrees because the question isn't new, has moved down the list and just isn't seen anymore.

- Late-comers who already answered a question in the comments section. In my language pair, for example, it is pretty common (unfortunately!) for questions to be discussed at length in the discussion box. Answerer A might give a good answer with the perfect explanation and relevant reference links in the discussion box without posting an actual answer. The discussion goes on and and at some later point answerer B posts a correct one-word answer which gets several agrees until someone points out that answerer A already said this earlier and should enter her answer so she can get the points. So we end up with two identical answers (one with agrees, one without) and the points should go to the answer without agrees (because that's the one where the answerer did all the legwork).

And as Lincoln and Heinrich already pointed out, this will not work for very specialized subject matters or rare language pairs where there might not be many users to agree or disagree with answers.

The decision about which answer is the most helpful should still be down to the user and not be dictated by some algorithm.

[Edited at 2015-01-05 09:32 GMT]
Collapse


 
Balasubramaniam L.
Balasubramaniam L.  Identity Verified
India
Local time: 05:56
Member (2006)
English to Hindi
+ ...
SITE LOCALIZER
kudoz is not designed to generate an accurate glossary Jan 5, 2015

In kudoz, that answer is selected which was most useful to the asker, and not necessarily the most accurate answer.

Also, as mentioned in my earlier post, in many kudoz forums such as in the English to Hindi kudoz, translators routinely subvert the rules to safeguard their own ranking in the kudoz list.

Furthermore, kudoz answers are graded by persons who are least qualified to grade them - the asker. In most cases, the asker has inferior linguistic abilities which is w
... See more
In kudoz, that answer is selected which was most useful to the asker, and not necessarily the most accurate answer.

Also, as mentioned in my earlier post, in many kudoz forums such as in the English to Hindi kudoz, translators routinely subvert the rules to safeguard their own ranking in the kudoz list.

Furthermore, kudoz answers are graded by persons who are least qualified to grade them - the asker. In most cases, the asker has inferior linguistic abilities which is why he is forced to ask in the kudoz forums. Only in very rare cases do competent translators resort to kudoz forums for clarification on really difficult linguistic or terminological issues. In 90% of the cases, the questions are such that they can be answered by referring to a good dictionary, or are such that no competent translator should ever have any doubts regarding them. (Again, this is based on my experience of the kudoz activity in my language pair).

So the persons grading the questions have no clue as to which is the best answer. It also needs to be kept in mind that kudoz are graded by taking into account how it would affect the translator ranking of the asker, the answerer and the agreers.

For all these reasons, kudoz is not geared to identify linguistically accurate answers.

One solution could be that a team of senior translators in all language pairs should carefully go through all the kudoz postings and vet them from the linguistic point of view. Only after this should entries go into the glossaries. Also, if while vetting, the team chooses a different answer for the glossary than the one chosen by the asker or the robot, then the answerer of the the chosen answer should be awarded the kudoz points and the kudoz points of the person who was selected earlier should be reduced.

This mechanism will ensure that the use of kudoz points for translator grading does not affect too much how translators respond to kudoz questions.
Collapse


 
Balasubramaniam L.
Balasubramaniam L.  Identity Verified
India
Local time: 05:56
Member (2006)
English to Hindi
+ ...
SITE LOCALIZER
More on agrees Jan 5, 2015

It has been my experience that people rush to put their agrees on the first few answers and rarely bother to come back to the question to review and agree with answers that come up later. Often, the latter answers are the better ones, for the simple reason that they would have the advantage of benefiting from the line of thinking in the earlier answers and would be in a position to correct errors or misjudgments in the earlier answers.

For this reason, it is generally observed that
... See more
It has been my experience that people rush to put their agrees on the first few answers and rarely bother to come back to the question to review and agree with answers that come up later. Often, the latter answers are the better ones, for the simple reason that they would have the advantage of benefiting from the line of thinking in the earlier answers and would be in a position to correct errors or misjudgments in the earlier answers.

For this reason, it is generally observed that the earlier answers have a string of agrees merely for the reason that they appeared first.

One way to correct this would be to put in place a reasonable time barrier before anyone can agree to answers. That is, agrees would be allowed only after say 3 hours of the question being posted. May be the site could do a statistical analysis of the various kudoz forums to identify the average time for each language pair in which the maximum number of answers to questions get posted and this could be the time barrier.

Another possible mechanism could be to allow translators to cancel their agrees after placing them. I think currently this is not possible. So if after agreeing to an answer, the question elicits another better answer, the agreer could cancel his agree on an earlier answer and place it on the new answer. Currently, all he can do is to also agree to the new answer. By allowing him, to cancel his earlier agree, we can enable him to fine-tune his agrees and pin-point the most correct answer, which I think would be more useful to the asker. This would also generate a more accurate kudoz glossary.
Collapse


 
Aakash5555
Aakash5555
India
Local time: 05:56
Member (2014)
English to Hindi
+ ...
I am totally agreed with you for English-Hindi Pair Jan 5, 2015

Mr. Balasubramaniam,
The points you raised about the Kudoz position in English-Hindi Pair are very relevant and there is a need to take some corrective measures regarding that. I totally agreed with you that for Translator's ranking Kudoz points tally should not be used.


 
Bernhard Sulzer
Bernhard Sulzer  Identity Verified
United States
Local time: 20:26
English to German
+ ...
TOPIC STARTER
no agrees = no Jan 5, 2015

Balasubramaniam L. wrote:

In kudoz, that answer is selected which was most useful to the asker, and not necessarily the most accurate answer.questions.


It should at least be accurate and, without agrees, there's no guarantee of any kind of how accurate or useful the answer is. If the asker wants to use it that's fine, but there shouldn't be points given and it should not get entered into the glossary. Instead, it should close automatically after a certain time if no one leaves any agrees or if it only garners neutrals or disagrees.


 
Andrea Halbritter
Andrea Halbritter  Identity Verified
France
Local time: 01:26
French to German
+ ...
Guarantees Jan 5, 2015

There are rarely guarantees in life...

2 peer agreements do not guarantee the correctness of an answer neither - just like an insurance policy won't guarantee that your house is not burning down one day...


 
Bernhard Sulzer
Bernhard Sulzer  Identity Verified
United States
Local time: 20:26
English to German
+ ...
TOPIC STARTER
couple of thoughts Jan 5, 2015

Andrea Halbritter wrote:

For the pair French to German it is often the same as for Chinese to English. So I agree with Lincoln Hui. There often are no agrees at all. Even peers that agreed in the discussion do not agree directly on the answers.


If the asker wants to use the answer, that's fine. And thanks to the lonely wolf who showed up and answered the question. But no thanks, I don't think there should be points given or terms entered into the glossary. Who says it's correct? The asker didn't know the answer, that's why he/she asked. He/she shouldn't be the one deciding that it is a valid answer without support by colleagues who agreed on the answer. The answerer might be right but he might be wrong.
What if no one else ever comments? How is it fair to award points to an answer that has not been supported? To say it didn't get any neutrals or disagrees wouldn't cut if for me either.

Andrea Halbritter wrote:
On the other hand one or two peer agrees do not assure the answer is correct neither. 3 or 4 may do but it also depends on if people who agree actually work in the field, if they are natives or not etc. Although there are a lot of really excellent answers it already happened to me as well to see answers with three peer agrees which turnt out to be false.


Well, let's just start somewhere. One agree is much better than no agree. But that's fine, I don't really have to participate. One less to agree or disagree.

Andrea Halbritter wrote:
So your proposition makes no sense to me and it is not possible to regulate everything. In the glossaries there will always be terms which are not correct. It's up to every single person who as a look at them to reverify.

[Modifié le 2015-01-05 06:49 GMT]


Well, I am sorry it makes no sense to you. It does make sense to me. And please read my initial description of the problem that prompted me to start this thread. In any case, thanks for commenting.


 
Pages in topic:   [1 2 3 4] >


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


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

Don't allow asker to select an answer which has not received any agrees or any comments at all






Wordfast Pro
Translation Memory Software for Any Platform

Exclusive discount for ProZ.com users! Save over 13% when purchasing Wordfast Pro through ProZ.com. Wordfast is the world's #1 provider of platform-independent Translation Memory software. Consistently ranked the most user-friendly and highest value

Buy now! »
TM-Town
Manage your TMs and Terms ... and boost your translation business

Are you ready for something fresh in the industry? TM-Town is a unique new site for you -- the freelance translator -- to store, manage and share translation memories (TMs) and glossaries...and potentially meet new clients on the basis of your prior work.

More info »