| Pages in topic: [1 2 3 4 5 6 7 8 9 10 11] > | Kudoz - Leader list - A good moment to sort by usefulness rate too? Thread poster: Tomás Cano Binder, BA, CT
|
|---|
Dear colleagues:
I am very aware of the fact that quick answers are sometimes preferred both by askers and by Proz.com themselves. But I also see that good answers are far more useful than quick answers, and that a majority of quick answers are not good enough and get ignored.
A quick calculation, Kudoz points / number of answers, reveals how useful has each person been in Kudoz. The Kudoz leader list can become a very good instrument, but it is not at present in my opi... See more Dear colleagues:
I am very aware of the fact that quick answers are sometimes preferred both by askers and by Proz.com themselves. But I also see that good answers are far more useful than quick answers, and that a majority of quick answers are not good enough and get ignored.
A quick calculation, Kudoz points / number of answers, reveals how useful has each person been in Kudoz. The Kudoz leader list can become a very good instrument, but it is not at present in my opinion, as now a critical factor is how long you have been in Proz. Reaching top positions in the list is very difficult for new users who provide very good help, so there is no incentive to do so.
So my proposal is to add a link to the Leaders list to have the option to sort the Top 50 list by usefulness, i.e. by the rate of Kudoz points / answers posted. This way:
A) When in doubt choosing an answer, askers will have a chance to see who of several answerers is more likely to have posted a solid answer. This is critical as sometimes you don't have the time to repeat your research and you need to be able to trust someone else.
B) New customers can choose wisely, based on the success rate (and thus reliability and/or knowledge) of the answerers, not on how many questions they answered or how long they have been Proz users.
C) More users will feel encouraged to help in Kudoz as they will have a chance to reach the top of the list, something they cannot do today as they can compete in knowledge and reliability, but not in time as users of Proz.com.
This would not mean removing the list as it exists now, but merely adding an alternative way of sorting it. I would also make it a bit longer (for instance 100 users) to encourage people to help in Kudoz and see how they progress along time.
Would you agree to such an alternative way of seeing the leaders list? ▲ Collapse
| | | | | Some comments | Jun 16, 2009 |
Hi Tomás,
I'm afraid this sorting method has been proposed many times before, to no avail. I guess we can have another go at it.
However, consider the following two points.
1) If a new user is any good in his/her field of specialization, it does not take long to accumulate enough points to be noticed.
2) Total points (over all fields) don't matter nearly as much.
That's why the system has worked well so far, and I believe it continues to... See more Hi Tomás,
I'm afraid this sorting method has been proposed many times before, to no avail. I guess we can have another go at it.
However, consider the following two points.
1) If a new user is any good in his/her field of specialization, it does not take long to accumulate enough points to be noticed.
2) Total points (over all fields) don't matter nearly as much.
That's why the system has worked well so far, and I believe it continues to do so.
[Edited at 2009-06-16 05:00 GMT] ▲ Collapse
| | | | | Reaching the top can take ages | Jun 16, 2009 |
Mikhail Kropotov wrote:
1) If a new user is any good in his/her field of specialization, it does not take long to accumulate enough points to be noticed.
Well, in the case of English to Russian it would take a new user 250 good answers just to be in the overall list, over 500 good answers to reach position 20, and 1,200 good answers to reach position 5. I think writing 250 good answers can take long enough!
In the case of Spanish you need over 500 good answers just to be in the overall list. It feels a bit like a gallery of old Proz.com users...
I agree that at least you should be able to write 250 good answers in English to Russian (or 500 in English to Spanish) to be in the overall list, but other than that, it should be possible to see who provides better answers.
Please notice that I am not suggesting the old list disappears. It's perfectly fine with me if it stays. I am only suggesting an alternative way of sorting the list so that it useful in more ways.
| | | | Aniello Scognamiglio (X) 
Germany
Local time: 12:53
English to German
+ ...
| Already in place! | Jun 16, 2009 |
Tomás Cano Binder, CT wrote:
...Reaching top positions in the list is very difficult for new users who provide very good help, so there is no incentive to do so...Top 50 list by usefulness...solid answer...success rate...reach the top of the list, something they cannot do today as they can compete in knowledge and reliability, but not in time as users of Proz.com.
Would you agree to such an alternative way of seeing the leaders list?
Good morning, Thomas,
what you are talking about is already in place and working fine. Let me make an example:
a) please go to the KudoZ menu,
b) choose the Leaders submenu
c) in the Select Language Pair dialog box select the language combo English-German (just an example) and click on the Go button
d) then, select IT in the Select specific category for detailed leaderboard: field
You should see 3 columns:
a) Last 3 months (I'm no 4)
b) Last 12 months (I'm no 1)
c) All Time (I'm no 1)
In my opinion, this categorization (3 months, 12 months, all time) introduced not long ago is a very good idea because it considers the time factor. What you want is already in place and working fine. Everybody can reach a top position in a specific field within only 3 or 12 months!
What irritates me slightly in your initial post are the terms "usefulness", "solid answers", "success rate" etc. I'm not a 100% sure what you want to say but in most cases the asker chooses the most useful answer. At least it's my experience in my primary language combo Eng-Ger. Did I miss something?
The current KudoZ point system is fair and transparent. There is no need to change it.
The alternative would be to get rid off the KudoZ point system altogether.
Even then how would you guarantee that the "best" answer is chosen?
Personally, I think too much importance is attributed to the KudoZ point system.
To be honest, I have earned a lot of KudoZ points in my specialist fields but they do not fill my refrigerator or pay my bills. However, colleages and potential clients can see at first glance in which fields they can trust me and expect a high level of quality.
Aniello
| | |
|
|
|
| Let me rephrase your last comment | Jun 16, 2009 |
To be honest, I have earned a lot of KudoZ points in my specialist fields but they do not fill my refrigerator or pay my bills. However, colleages and potential clients can see at first glance in which fields they can trust me and expect a high level of quality.
Aniello [/quote]
Dear Aniello,
Regarding the last part, I partly disagree. Let me rephrase it. I believe that KudoZ points in your specialist fields (mine too) DO fill your refrigerator or pay your ... See more To be honest, I have earned a lot of KudoZ points in my specialist fields but they do not fill my refrigerator or pay my bills. However, colleages and potential clients can see at first glance in which fields they can trust me and expect a high level of quality.
Aniello [/quote]
Dear Aniello,
Regarding the last part, I partly disagree. Let me rephrase it. I believe that KudoZ points in your specialist fields (mine too) DO fill your refrigerator or pay your bills indirectly, i.e. by looking at one's earned KudoZ points in specific fields will in the future and have in the past lead to and led to customers, except that you haven't noticed it. ▲ Collapse
| | | | Erik Freitag
Germany
Local time: 12:53
Member (2006)
Dutch to German
+ ...
Aniello Scognamiglio wrote:
What irritates me slightly in your initial post are the terms "usefulness", "solid answers", "success rate" etc. I'm not a 100% sure what you want to say but in most cases the asker chooses the most useful answer. At least it's my experience in my primary language combo Eng-Ger. Did I miss something?
If I may step in here: If I understand Tomás' suggestion correctly, what he means by "usefulness", "solid answers", "success rate" is: If a Kudoz participants posts an answer, how likely is it going to be chosen as the most helpful one? This question can be answered in terms of the questions answered/points earned ratio. This ratio in a way reflects the overall quality of the participant's answers, though of course it doesn't say anything about the quality of an individual answer.
In the EN-DE language pair for example, there is one participant who very actively contributes. A lot of questions are answered, obviously mostly quick shots without any grasp of the subject, and the quality of the language is always absolutely substandard. However, out of the last 50 questions he answered, three were chosen as the most helpful one, which earned him a total of 6 Kudoz points for these 50 questions. Due to the massive amount of questions answered, he still is amongst the top 5 in the Kudoz ranking list for the last 3 and 12 months. A potential customer might get the impression that this person is a very good translator, while in fact any answer he posts is most likely utter nonsense. All that glistens is not gold.
I find Tomás' suggestion very good and well worth implementing (and I have said so before in similar threads). One thing to consider: A minimum number of questions (maybe 50?) should be necessary before a participant appears in said list. Otherwise the best strategy would be: 1. Answer one question where you're absolutely sure that you'll get the points. 2. Answer no more questions. That way, your ratio will be 100% (with only one question answered), and you'll be (and stay) right on top of the list.
My 2 cents,
regards,
Erik
| | | | | Surely in all pairs | Jun 16, 2009 |
efreitag wrote:
In the EN-DE language pair for example, there is one participant who very actively contributes. A lot of questions are answered, obviously mostly quick shots without any grasp of the subject, and the quality of the language is always absolutely substandard. However, out of the last 50 questions he answered, three were chosen as the most helpful one, which earned him a total of 6 Kudoz points for these 50 questions.
Exactly. Actually my main point is that good help (or useful help, or solid help) DOES EXIST in Kudoz, but is rarely recognized by the system. Personally I only answer questions if I am pretty sure that I am being useful, although I could answer at random as the next person!
Let's add the other sorting order to give that little bit of recognition to the other way of helping in Kudoz!
| | | | | I did not mean to change anything!! | Jun 16, 2009 |
Aniello Scognamiglio wrote:
The current KudoZ point system is fair and transparent. There is no need to change it.
But Aniello, I did not mean to change Kudoz, the point system, anything! That all has been discussed to exhaustion ever since Kudoz exists. I did not mean to change anything.
I just propose to ADD the OPTION to see the leader list in a different way that can be more meaningful for some uses of that information.
| | |
|
|
|
efreitag wrote:
I find Tomás' suggestion very good and well worth implementing (and I have said so before in similar threads). One thing to consider: A minimum number of questions (maybe 50?) should be necessary before a participant appears in said list. Otherwise the best strategy would be: 1. Answer one question where you're absolutely sure that you'll get the points. 2. Answer no more questions. That way, your ratio will be 100% (with only one question answered), and you'll be (and stay) right on top of the list.
Well, of course the idea is that the new list would only contain the same people who make the current Top 50 in each of the three periods. That way you make sure that people keep contributing actively.
| | | |
I like Tomas suggestion, but as Mikhail says, it has already been suggested several times with no results.
efreitag wrote:
In the EN-DE language pair for example, there is one participant who very actively contributes. A lot of questions are answered, obviously mostly quick shots without any grasp of the subject, and the quality of the language is always absolutely substandard. However, out of the last 50 questions he answered, three were chosen as the most helpful one, which earned him a total of 6 Kudoz points for these 50 questions. Due to the massive amount of questions answered, he still is amongst the top 5 in the Kudoz ranking list for the last 3 and 12 months.
Same in my pair!
One thing to consider: A minimum number of questions (maybe 50?) should be necessary before a participant appears in said list. Otherwise the best strategy would be: 1. Answer one question where you're absolutely sure that you'll get the points. 2. Answer no more questions. That way, your ratio will be 100% (with only one question answered), and you'll be (and stay) right on top of the list.
Agree too.
Stéphanie
| | | | | Opinions so far: 3 for it; 2 against it. | Jun 16, 2009 |
Thanks guys! Share your opinion, even if it is just a For/Against! Cheers!
| | | | | How about acceptance rate instead? | Jun 16, 2009 |
The number of points awarded for a successful answer can vary so why not consider using the acceptance rate instead? This figure is available for members in the Kudoz stats and is independent of the number of points awarded by the asker. In my opinion this could help to identify those people who take the "shotgun approach" to answering questions, i.e. answer everything in sight and then go away and do some research to justify their answer, which often turns out to be wrong, so they then give a d... See more The number of points awarded for a successful answer can vary so why not consider using the acceptance rate instead? This figure is available for members in the Kudoz stats and is independent of the number of points awarded by the asker. In my opinion this could help to identify those people who take the "shotgun approach" to answering questions, i.e. answer everything in sight and then go away and do some research to justify their answer, which often turns out to be wrong, so they then give a different answer in their third or fourth additional comment.
Andy
FYI My acceptance rate in Es > En is 53%. ▲ Collapse
| | |
|
|
|
| Yes, of course | Jun 16, 2009 |
Dr. Andrew Frankland wrote:
The number of points awarded for a successful answer can vary so why not consider using the acceptance rate instead? This figure is available for members in the Kudoz stats and is independent of the number of points awarded by the asker. In my opinion this could help to identify those people who take the "shotgun approach" to answering questions, i.e. answer everything in sight and then go away and do some research to justify their answer, which often turns out to be wrong, so they then give a different answer in their third or fourth additional comment.
Yes, that also makes sense. However, I think it might make things more complicated on the development side. I don't know! Thanks for sharing!
| | | | Erik Freitag
Germany
Local time: 12:53
Member (2006)
Dutch to German
+ ...
| Why not make statistics accessible to the public? | Jun 16, 2009 |
As a side remark: Why is the Kudoz statistics page hidden to the public? Wouldn't it be a good idea to allow others access to the information, and expose that page more prominently?
| | | | | Acceptance rate | Jun 16, 2009 |
I would love to see answerers listed by acceptance rate.
But efreitag has a point.
Somehow it has to be possible to place someone with 250 accepted answers higher than the person with only 5, even if both have the same acceptance rate.
Personally I can maintain my position between 3 and 5 in the 'last 3 months' list for my main pair with far less effort than I used to spend on Kudoz. I suspect some of my colleagues lower down the list have higher acceptance rates, but t... See more I would love to see answerers listed by acceptance rate.
But efreitag has a point.
Somehow it has to be possible to place someone with 250 accepted answers higher than the person with only 5, even if both have the same acceptance rate.
Personally I can maintain my position between 3 and 5 in the 'last 3 months' list for my main pair with far less effort than I used to spend on Kudoz. I suspect some of my colleagues lower down the list have higher acceptance rates, but they do not answer nearly so many questions. (My overall average is just over 50%).
I wonder if it would be an idea to highlight KudoZ editors (users who have been given the right to perform some editing functions on KudoZ questions).
I was among those who suggested changing the procedure for selecting KudoZ editors.
As Moderator I wanted to nominate one answerer who must have had an acceptance rate around 80% or more, but she had barely 300 points after 5 years of membership... and would take years to get to the 500 level. She only answered in a very narrow field, but when she did answer, she was absolutely reliable.
I did finally persuade staff to invite her to be an editor, so that she could tidy up the glossary a little in her field. People like that are the very best contributors to KudoZ, but do not always make the top of the list.
The minimum number of points for editors is now 300, but candidates have to be proposed and invited, and this does not happen automatically.
There is simply not that much traffic in some narrow fields, but in 'small' languages there are not many resources either, and KudoZ, with its chance for discussion, is an important tool.
I would love to see the emphasis shifted from quantity towards quality. Some kind of weighting for acceptance rate might encourage specialists in narrow subject areas, and anyone who makes occasional, but really valuable contributions.
▲ Collapse
| | | | | Pages in topic: [1 2 3 4 5 6 7 8 9 10 11] > | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Kudoz - Leader list - A good moment to sort by usefulness rate too? | 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 » |
| | 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! » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
