| Pages in topic: < [1 2 3 4] > | KudoZ grading system needs to be changed! Thread poster: sarandor
|
|---|
Eric Hahn (X) 
France
Local time: 18:02
French to German
+ ...
Stéphanie Soudais wrote:
The remedy is also for answerers not to answer wildly before having more context (BTW it would also educate the asker)... But some just can't help answering and trying their luck to gain points.
To resume it:
It's a good practice to ask questions when you are doubtful.
But to give doutbtful answers is not good for your image, too
[Edited at 2009-02-14 16:21 GMT]
| | | | Gina W
United States
Local time: 12:02
Member (2003)
French to English
| The best answer should be chosen by the Asker | Feb 14, 2009 |
Birgit Richter wrote:
"- Let the best answer be chosen not by the asker, but by peers (it will also rid askers from the unpleasant task of choosing the best answer)"
I disagree entirely. ONLY the asker is in a position to decide which answer is the most helpful for their particular translation problem as they know the original text in its entirety - the answerers don't.
The Asker has the text. Also, the Asker IS a peer, not some clueless person posting a question which s/he has no idea about. Plus, the peer agreement is often a result of an Answerer's "popularity" on the site - not on whether or not the answer is the best one.
As for the question of context, an Asker can post context but can not possibly post ALL of the context - meaning you can't post the whole text. I mean I guess someone could but I certainly would not recommend it nor would I think it should be considered absolutely necessary.
| | | | Gina W
United States
Local time: 12:02
Member (2003)
French to English
| Yes, we do need to close a question without grading | Feb 14, 2009 |
kdeimling wrote: Steven Capsuto wrote:
No, please keep that feature!
If I ask a question and get three answers that are patently wrong, or if there are three answers but none is well documented and I don't know which, if any, is correct... what do you propose? That I roll dice to decide who gets the points? Usually, these are cases where none of the responses has "agrees" or "disagrees." There isn't much to go on.
After a week or two, I just close the question. It seems the only fair option.
[Edited at 2009-02-13 20:05 GMT] I agree with Steven. I've been in the same situation, though it doesn't happen often, and I don't see how we can grant points to answers that we realize are not correct!
I haven't done it often but don't see any reason that feature should be done away with. I don't think it's always possible for the Asker to have to justify this, though. Particularly because what if the Asker would prefer the question be removed, yet the site only allows Editors to remove questions? So because the Asker is now stuck with the question on the site forever, s/he HAS TO give points, whether or not the answers are helpful, and explain why? While in most cases an Asker can, there may be some cases when an Asker can not.
So many changes have been made to the KudoZ grading system already, btw. And most of those changes have benefitted everyone BUT the Asker.
If you're going to make another change to the system, it should be that an Asker should be able to block out certain Answerers. If certain individuals have absolutely nothing constructive to add, it would save an Asker time and aggravation not having to read that person's comments. JMO.
| | | | | Melzie, you've read my mind! | Feb 14, 2009 |
Melzie wrote:
I have a question for you all.
When you use the KudoZ term search, do you automatically pick the answer chosen by the original answerer or do you go through all the answers to see which suits your present translation best?
Asking and answering a Kudoz question - and maybe even giving and getting points - is only the beginning of the process. The answers won't just help the original asker, OK getting points is nice but it's not going to be life changing in many cases, they are on record for the community that we form.
My modification to the system would be to add a box after each answer for a searcher to click with a "Thank you, your answer was the most helpful to me today." maybe many years after the original question.
A "peer thanks" column could be added on the points page, giving more insight on a translator.
I have the intention of saying precisely this on the forum, except you've said it much better.
Answer to your question: no, I do not automatically choose the answer chosen by the original asker. And many times, I found good, thoughtful, well researched and well documented answers which were totally ignored by askers.
As "peer thanks" column does not exist yet (and it would be a great idea!), I made it a point to use the "agree" option on old answers (in some cases, 5 or 6 years old) to thank the person who provided the answer helpful to me, especially when it was not originally appreciated. I just want that answerer to know that he/she wasn't wasting his or her time.
| | |
|
|
|
1. I positively hate the task of choosing! When two or more people gave me the most precious thing a person has - their time, when they took the trouble to research the question and provide references and/or detailed explanation, and I have to reward only one of them - honestly, I don't feel good about it.
2. The option of closing without grading should exist - but askers should be using it very, very sparingly. I've seen many cases where several good answers existed, but asker... See more 1. I positively hate the task of choosing! When two or more people gave me the most precious thing a person has - their time, when they took the trouble to research the question and provide references and/or detailed explanation, and I have to reward only one of them - honestly, I don't feel good about it.
2. The option of closing without grading should exist - but askers should be using it very, very sparingly. I've seen many cases where several good answers existed, but asker chose to close without any grading or explanation.
3. I think it's OK to asnwer when you are in doubt, as long as you choose confidence level 1 (wasn't that level specifically created for that purpose?) ▲ Collapse
| | | | Zea_Mays
Italy
Local time: 18:02
Member (2009)
English to German
+ ...
Paul Cohen wrote:
(...)
What's more, this site doesn't allow outsiders to search for translators according to their success rates. The directory can be sorted according to language combination, native language, place of residence, PRO Certification, etc., and perhaps most importantly, the sheer number of KudoZ pointZ, but not according to a person's KudoZ success rate, i.e., the number of pointZ divided by the number of answers provided.
Come to think of it, a "KudoZ success rate" directory search feature would help erase the advantage that older, somewhat trigger-happy members with mountains of pointZ have over newcomers. I wonder if there have been any discussions of this possibility in the past? What do staff members think?
In my opinion, we don't need to change the grading system. We just need to make it easier for people to interpret what the pointZ actually mean. I totally agree.
The "success rate" should be shown beside the answers and/or points in the ranking lists and on profile pages, let's say after the first 3-4 months or from xxx points upwards.
From my point of view, there should be NO option to hide these statistics.
Relating to this, I've seen an interesting but expressive dynamic: people with thousands of provided answers but LOW "succes rate" HIDE their statistics.
It's sometimes the same people who appears ON TOP in the rankings, since the system regards only the POINTS!
I disagree also with the idea to let the best answer be choosen by peers, since I've seen often happens (already) exactly this - when the asker chooses the answer with the most agrees, although it is completely wrong.
More often I've seen askers choosing the right answer, although there were others with a lot more agrees.
| | | | Stephen Rifkind
Israel
Local time: 19:02
Member (2004)
French to English
+ ...
| Why and I how I use kudoZ | Feb 15, 2009 |
I cannot speak for others, but I pose questions to kudoZ only after checking all of my dictionaries, online sources, including the Proz search, and googling the term. Often, the problem is caused by a rare use of a word. Yes, I provide as much context as I can. So, translations that are correct in most cases may not be appropriate in this specific case. Also, I may choose to use a more British or American term based on the text in hand.
The asker is required to pick the most hel... See more I cannot speak for others, but I pose questions to kudoZ only after checking all of my dictionaries, online sources, including the Proz search, and googling the term. Often, the problem is caused by a rare use of a word. Yes, I provide as much context as I can. So, translations that are correct in most cases may not be appropriate in this specific case. Also, I may choose to use a more British or American term based on the text in hand.
The asker is required to pick the most helpful term. It is worthwhile to view the discussion if someone wants to see other translations of a term.
All this being so, peers cannot choose the helpful answer. Nor are the people asking always lazy or non-proficient in that language. ▲ Collapse
| | | | sarandor
United States
Local time: 12:02
English to Russian
+ ...
TOPIC STARTER | Just a few more comments... | Feb 15, 2009 |
Thank you all for sharing your thoughts on this issue. I agree that it is impossible to design a system that would make everyone happy.
Here are just a few more comments that I would like to add to the discussion:
- I agree with colleagues who pointed out that people who ask questions are not necessarily lazy or unqualified. I had to train new employees for one of the jobs I held in the past before becoming a freelancer. I've noticed that some people would NEVER, NEVER... See more Thank you all for sharing your thoughts on this issue. I agree that it is impossible to design a system that would make everyone happy.
Here are just a few more comments that I would like to add to the discussion:
- I agree with colleagues who pointed out that people who ask questions are not necessarily lazy or unqualified. I had to train new employees for one of the jobs I held in the past before becoming a freelancer. I've noticed that some people would NEVER, NEVER ask questions because they are afraid to look stupid, and these are the people who make the most stupidest of mistakes, excuse my English.
- The current system rewards the answerer who provides the most helpful answer from the asker's point of view. Which would've been fine, if the points were not used as an indicator of one's quality as a translator. The most helpful answer for asker is not the same thing as the best answer IN THE CONTEXT PROVIDED (or information given). The latter is a far more objective measure of one's abilities as a translator. I think that not displaying the number of questions answered would to some extent solve the problem of subjectiveness.
- Some colleagues expressed an opinion that KudoZ and its points should not be taken too seriously. Well, define "too seriously"... If something affects my business, I take it seriously. KudoZ helps me to provide better translations for my clients, and the points and other KudoZ-related information are displayed on quotes that I submit and on my profile page. At least in theory it can influence outsources' decisions whether to hire me or not.
But thank you again everyone, and I wish a good productive week to you all! ▲ Collapse
| | |
|
|
|
| Most helpful answer | Feb 16, 2009 |
Eric Hahn wrote: OK, then let's put the question differently Enrique wrote:
The asker selects the answer that he/she found most helpful, not the "best translation".
How does he know which answer is the most helpful ?
Who will be better qualified than the asker to know which answer was more helpful to him/her?
Regards,
Enrique
| | | | | KudoZ includes 'confidence levels' - why? | Feb 16, 2009 |
I find it fascinating dipping into KudoZ. I post an answer when I see that other suggestions (if any) either haven't caught the correct meaning, or I think an alternative translation may be relevant.
Often there is not enough context in the question - and it is an option to ask for further information, remember - so sometimes I will comment in my answer that if the context is such and such, then I think my suggested translation fits the best, but not in another context. Or my transl... See more I find it fascinating dipping into KudoZ. I post an answer when I see that other suggestions (if any) either haven't caught the correct meaning, or I think an alternative translation may be relevant.
Often there is not enough context in the question - and it is an option to ask for further information, remember - so sometimes I will comment in my answer that if the context is such and such, then I think my suggested translation fits the best, but not in another context. Or my translation might be best if the document is more (or less) formal.
In these cases I'm not worried if another answer suits the asker better. I'm just trying to be helpful, while exercising my translating brain at the same time. It is good practice.
Maybe I should be more concerned about my KudoZ points.
Another issue is that we can indicate 'confidence levels'.
I rarely answer if my confidence is not 'high'. Sometimes I'll give a 'medium' ranking, because I am making some assumptions that may not be valid.
Again, when people are giving lower confidence rankings, it is surely because they are trying to be helpful.
I think askers often look at the proposed answers and they are a great prompt, either jogging their memories or showing a research direction enabling them to confirm the best words to use. For them it is useful having suggestions that are not tagged 'I'm certain' and only posted by people who want to win 4 points or a glossary entry.
So my view is KudoZ is primarily to help other translators and to practice my own skills - it is a useful investment of my time if I need a breather or don't have other work at that moment.
I would hope that anyone who uses KudoZ points would actually look at a page or two of the table that you can access with one click from someone's profile, showing their suggested translations. I stand by every translation I've offered and think they are a credit to my abilities even if they didn't receive 4 points.
However, I suspect KudoZ points are of most relevance in the ranking they give in searches of the translator database. The more points you have, the higher you are on the list. But that is really a measure of your participation in the community as it seems to be the simple total that is used, rather than a ratio of chosen to not-chosen answers.
It would be interesting to hear from outsourcers if and how they use KudoZ points/entries.
[Edited at 2009-02-16 13:05 GMT] ▲ Collapse
| | | | Anna Fominykh
United States
Local time: 11:02
English to Russian
+ ...
| Thank you for raising this topic, Galia! | Mar 18, 2009 |
For some time now I wanted to raise this topic myself, and Galia put in words precisely and beautifully. I was referred to this forum only today, after similar discussion had arisen after the Asker choose not the most correct answer (but may be most helpful for her).
I agree with so many people, who I am quoting below. They all have named the valid reasons for the system to be modified to some extent.
ElleC wrote: Tim Drayton wrote: voyager wrote:
- Let the best answer be chosen not by the asker, but by peers (it will also rid askers from the unpleasant task of choosing the best answer)
We all get stuck from time to time, but let us be honest. The people who are most likely to ask KudoZ questions are those with a shaky knowledge of terminology in the field in question and who, furthermore, are not very good at researching. Such people may well not be the best judge of the answers. I have occasionally seen cases in my own language pair where people have been incapable of recognizing the right answer when it is staring them in the face and backed up with credible sources. How does the questioner really know that it is the best answer? Aren't they asking the question initially because they're not sure? I recently answered a question and received the majority of agreements. However, two days later, someone came in with an incorrect answer and they got the points! They used an uncommonly used word, but for some reason the questioner chose them. The answers should be graded automatically. How do I know the questioner didn't decide to reward a friend?
Enrique wrote:
The asker selects the answer that he/she found most helpful, not the "best translation".
How does he know which answer is the most helpful ?
But the problem remains the same: Bad answers can't be helpful.
Especially, when they end up in the glossary !
gad wrote:
The Asker has the text. Also, the Asker IS a peer, not some clueless person posting a question which s/he has no idea about. Plus, the peer agreement is often a result of an Answerer's "popularity" on the site - not on whether or not the answer is the best one.
My main concern apart from preventing new members earning well deserved points, is that not correct, and sometimes completely opposite to being correct answers are chosen and being added to the Glossary! I find this beyond any excuse. It discredits the site, it harms us all, and makes the points earned worth less (not completely worthless, I hope:))
The best thing in this case, as Enrique recommends, is to contact the moderators, and I can say that I did. And one moderator in our EN-RU language pair after reviewing my suggestion did changed the glossary entry, as previously accepted entry with 4 points awarded was wrong.
My suggestion is, that one very simple thing to be added to the existing system: after the Asker chooses the most helpful answer it HAS TO BE REVIEWED BY MODERATOR, and then points could be awarded and a glossary entry made. This would add some control to the situation and hopefully some fairness to grading, provided, of course, that moderators would be a third party person, not personally interested in the outcome. And for this extra work moderators could be awarded extra points or distinguished in some other way.
P.S. There should be some kind of "suggestion box" on this site, where these kind of suggestions could be submitted. Or is it forums that serve it purpose here?
[Edited at 2009-03-19 16:04 GMT]
| | | | | Clarifications | Mar 18, 2009 |
Anna Fominykh wrote: Enrique wrote:
The asker selects the answer that he/she found most helpful, not the "best translation".
How does he know which answer is the most helpful ? But the problem remains the same: Bad answers can't be helpful. Especially, when they end up in the glossary ! gad wrote:
The Asker has the text. Also, the Asker IS a peer, not some clueless person posting a question which s/he has no idea about. Plus, the peer agreement is often a result of an Answerer's "popularity" on the site - not on whether or not the answer is the best one. My main concern apart from preventing new members earning well deserved points, is that not correct, and sometimes completely opposite to being correct answers are chosen and being added to the Glossary! I find this beyond any excuse. It discredits the site, it harms us all, and makes the points earned worth less (not completely worthless, I hope:)) The best thing in this case, as Enrique recommends, is to contact the moderators, and I can say that I did.
Please note that I suggested contacting a moderator if you suspect any wrongdoing (i.e. rules infringement), not if you disagree with the answer chosen by the asker.
My suggestion is, that one very simple thing to be added to the existing system: after the Asker chooses the most helpful answer it HAS TO BE REVIEWED BY MODERATOR, and then points could be awarded and a glossary entry made.
The document describing the role of moderators states: "However, it should be understood that moderators do not serve to any degree as linguistic authorities."
Regards,
Enrique
| | |
|
|
|
ElleC
United States
Local time: 11:02
English to Spanish
+ ...
| Not always the asker... | Mar 18, 2009 |
Enrique wrote:
Who will be better qualified than the asker to know which answer was more helpful to him/her?
Regards,
Enrique
Many times the Asker is not translating into his/her native language, so they might not be aware that the term they picked is incorrect.
| | | | | Most useful answers | Mar 19, 2009 |
ElleC wrote: Not always the asker... Enrique wrote:
Who will be better qualified than the asker to know which answer was more helpful to him/her?
Regards,
Enrique Many times the Asker is not translating into his/her native language, so they might not be aware that the term they picked is incorrect.
The askers are the ones who have the right to decide what answers were perceived by them as the most useful.
Regards,
Enrique
| | | | | KudoZ is there to benefit askers | Mar 19, 2009 |
Enrique wrote:
The askers are the ones who have the right to decide what answers were perceived by them as the most useful.
Regards,
Enrique
I'm with Enrique.
What does it say at the top of the KudoZ listing? "The KudoZ network provides a framework for translators and others to assist each other with translations or explanations of terms and short phrases."
While I understand the desire to build up Kudoz points, that is not the primary purpose of the network, nor, in my view, should it be the primary purpose for posting answers.
There are plenty of tools on the board for providing comments on answers and to have a discussion about the context. When I search the database I look at those as much as the selected answer. I quite often see people telling the asker they have selected the wrong answer. I think these checks are sufficient safeguard. While askers sometimes select a wrong answer, I've also seen wrong answers have most peer support.
| | | | | 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 » KudoZ grading system needs to be changed! | 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! » |
| | 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 
