Pages in topic: [1 2 3] > |
Slowing down the KudoZ process Thread poster: IanW (X)
|
IanW (X) Local time: 10:53 German to English + ...
Dear all, In a recent posting by Dusty www.proz.com/topic/32179?start=60, Kim Metzger brought up a very interesting point: […] I also feel very strongly about the need to do something to slow down the KudoZ process a bit. As it stands now, the perception that getting in an answer before anyone else has a chance to is positively harmful.www.proz.com/topic/32179?start=60, Kim Metzger brought up a very interesting point: […] I also feel very strongly about the need to do something to slow down the KudoZ process a bit. As it stands now, the perception that getting in an answer before anyone else has a chance to is positively harmful. I’m an active contributor on the German-English and French-English pairings, where, in the case of straightforward questions, there is generally a mad scramble to get the answer in first (frequently, I might add, with my good self racing with the pack as well). Granted, the asker does get a choice of four or five answers in the time it takes to make a coffee, but these often come without explanations, riddled with typos and often without the answerers having read the question in full and the context provided. In short, the quality of the answers is often negligible. If the time pressure were taken away, answerers who wished to do so would be able to take their time and research their answers, backing them up with suitable examples and “Dusty-evskian” explanations. This would mean slim pickings for point-grabbers (particularly those who tend to stray too far from home) and an end to the dominance of broadband users and – I firmly believe – would lead to a much-needed increase in the quality in KudoZ answers. How about introducing a system whereby all answers received within the first hour are marked first hour rather than 1 minute, 2 minutes etc. and whereby all answers are displayed in a random order. Askers would continue to receive the answers as they are submitted, but are encouraged to award points to the best answer rather than the fastest one. As well as assisting the asker in question, this would make KudoZ questions a far more valuable tool for the countless translators who refer to them long after the points have been awarded. This is just a rough idea for the moment, but I hope that, with input from my colleagues, it can be developed into something capable of bringing about a change for the better. Any suggestions, comments or critiques – with the exception of knee-jerk reactions from compulsive point-grabbers – would be very much appreciated. All the best Ian ▲ Collapse | | |
Silvia Montufo Urquízar (X) Spain Local time: 10:53 French to Spanish + ... Great suggestion | May 15, 2005 |
Very quickly, I am having some surgery tomorrow and must go to bed in a minute... I am totally for the system you are suggesting. Answering questions just shouldn't be a race and the quality of the answers, which are there for everyone to use, will be better. Best regards, Silvia | | |
24-hour hold | May 15, 2005 |
Blocking the ability to close a question within the first 24 hours after it is asked would be another essential ingredient of a system to improve quality by reducing urgency. This measure has been proposed many times and discussed extensively (http://www.proz.com/topic/31256, for a recent example).
[Edited at 2005-05-15 22:30] | | |
Points chasers & hide agrees & 24 hours | May 15, 2005 |
I see one potential flaw. As things stand, identical or almost identical postings are at least frowned upon by the more rational & reasonable kudoz users (small time delays notwithstanding). If the time displayed was just "within 1st hour", there would be no brake at all on the inveterate points chaser posting an identical answer, adding a quick note or web ref to make it look better and thereby attracting all the agrees, and therefore more likely to be chosen and therefore scoopin... See more I see one potential flaw. As things stand, identical or almost identical postings are at least frowned upon by the more rational & reasonable kudoz users (small time delays notwithstanding). If the time displayed was just "within 1st hour", there would be no brake at all on the inveterate points chaser posting an identical answer, adding a quick note or web ref to make it look better and thereby attracting all the agrees, and therefore more likely to be chosen and therefore scooping the points they crave. The type of questions & answers where what you describe tends to happen most are those (as I think you've said, I can't see it at the moment )where the question is fairly straightforward, as are the answers, the agrees pile in, and the answer with most agrees is selected 10 minutes later. Rather than your suggestion (I see nothing intrinsically wrong with precise timing and don't think this is in itself the cause of the problem you describe), I think this could be avoided if all peer comments were hidden for the first hour. And in addition, to help the quality process as discussed ad nauseum and which no-one seems to have a problem with per se, to not "kindly ask" but to absolutely block askers from selecting an answer until 24 hours have passed.
[Edited at 2005-05-15 23:10] ▲ Collapse | |
|
|
------ (X) German to Spanish + ... Need for speed | May 16, 2005 |
I basically agree with your suggestions, although I personally enjoy answering kudoz questions as fast as I can. On the english>spanish pairing there are at least six prozians answering correctly within the first 55 seconds and that is big fun. Of course, the quality would be probably better after one hour but most of the players/translators are using the add content option. Speed is very important in our business with usually short deadlines and customers who come around with 5000 words "urgent... See more I basically agree with your suggestions, although I personally enjoy answering kudoz questions as fast as I can. On the english>spanish pairing there are at least six prozians answering correctly within the first 55 seconds and that is big fun. Of course, the quality would be probably better after one hour but most of the players/translators are using the add content option. Speed is very important in our business with usually short deadlines and customers who come around with 5000 words "urgently until tomorrow". Also broadband connections are very important, not only for earning kudoz points but for translators in general. I would like to add that I do not agree with the system of pro and non-pro questions. In my opinion it is often impossible to decide if a question is easy or not easy and it is senseless to make the difference because there are hundreds of non-pro questions with 4 very different answers each. And how can the difference between answers be so important if the question is supposed to be easy? For the same reason the actual ranking without counting non-pro points, which personally benefits me, is not very fair. I finally want to suggest that it possibly could be a positive change to answer questions anonymously. Best regards ▲ Collapse | | |
Slowing down the KudoZ process | May 16, 2005 |
Janfri wrote: I basically agree with your suggestions, although I personally enjoy answering kudoz questions as fast as I can. On the English>Spanish pairing there are at least six prozians answering correctly within the first 55 seconds and that is big fun. Hi Janfri, yes, they used to call me Fast Eddy and Speedy Carnicero. It's great fun to find a great translation with a great reference in one or two minutes (and learning how to research KudoZ questions quickly has helped me enormously in my translation work. And your point about translators sometimes needing a translation quickly is well taken. We've all been there - and maybe we could have an exception in those cases. Questions marked "urgent" could bypass Ian's proposed waiting period. But I'd be willing to forgo the thrill of seeing a well-researched answer entered within a few minutes if it meant improving the overall quality of KudoZ. | | |
The system is balanced | May 16, 2005 |
I think the system is balanced as it is. Not perfectly balanced by any means, but to put it simply, it works! Charlie's suggestion looks interesting, i.e. hiding agrees for the first hour. It might be a viable solution to imperfect, incomplete answers getting lots of agrees and often hasty grades. To my taste though this time might be worth cutting to 20-30 minutes. An idea that has popped up from time to time in my pairs was to display seconds together with the minute... See more I think the system is balanced as it is. Not perfectly balanced by any means, but to put it simply, it works! Charlie's suggestion looks interesting, i.e. hiding agrees for the first hour. It might be a viable solution to imperfect, incomplete answers getting lots of agrees and often hasty grades. To my taste though this time might be worth cutting to 20-30 minutes. An idea that has popped up from time to time in my pairs was to display seconds together with the minutes. This is due to two or more similar or even identical answers all showing up at 2 minutes or even at 1 minute. At that point, adding more content to one's answer penalizes the author since the answer automatically drops to the bottom of the page, making it appear to have come in later than it did. A simple addition of seconds to the time format would prevent a lot of bickering about who was first. Now, this might be going right against what Ian is saying, but that's how it looks from my perspective. Mike P.S. Sorry, I meant Charlie's suggestion! Corrected.
[Edited at 2005-05-16 06:32] ▲ Collapse | | |
Kirill Semenov Ukraine Local time: 11:53 Member (2004) English to Russian + ... Opens ways for kudoZ hunters | May 16, 2005 |
Charlie Bavington wrote: I see one potential flaw. As things stand, identical or almost identical postings are at least frowned upon by the more rational & reasonable kudoz users (small time delays notwithstanding). If the time displayed was just "within 1st hour", there would be no brake at all on the inveterate points chaser posting an identical answer, adding a quick note or web ref to make it look better and thereby attracting all the agrees, and therefore more likely to be chosen and therefore scooping the points they crave. This was exactly my first thought. Many kudoz hunters' approach is to steal others' answers or give slightly edited versions based on other answerers' suggestions. As you rightfully noted, implementing of Ian's suggestion may lead to a lot of almost identical answers (with the only difference being whose answer is `longer' or contains a huger `tail' of Google references). As for broadband connection, I would not over-estimate its importance. I used to have a dial-up one until recently, and it was also OK. Both typing technique and a good skill of Googling mentioned by Kim are much more important. In many cases 2 or 3 minutes are enough to find a nice answer (in most cases it means that the asker hasn't tried to perform any research before asking his or her question, but this is another story...) | |
|
|
Mats Wiman Sweden Local time: 10:53 Member (2000) German to Swedish + ... In memoriam The KudoZ system is for the asker, not the answerer | May 16, 2005 |
The asker often is in stress. In that situation, hiding the Agree is detrimental. Seeing who is agreeing is soften instrumental for the asker's confidence in an answer. A knowledgeable answerer's answer in the field at hand often suffices. Anonymous answers have no such 'quality tag' attached to them. I agree with Mikhail Kropotov: "The system is balanced" Mats | | |
Linn Arvidsson (X) Sweden Local time: 10:53 English to Swedish + ... I agree with Mats | May 16, 2005 |
Seeing who is agreeing is soften instrumental for the asker's confidence in an answer. That is very true and something I often use. | | |
But, the asker is likely to benefit from more thorough answers | May 16, 2005 |
Dear Mats, Let me take another side's point of view for a moment here. If the time factor were to be lowered, the asker would be one to benefit. Getting hasty and shallow answers is only detrimental to the process of discovering the best possible answer. Mike | | |
Sonja Tomaskovic (X) Germany Local time: 10:53 English to German + ...
There is always some discussion on the KudoZ system and how to revise it or make it better. I cannot remember one single discussion where people would not start debating about the pros and cons of this and that or how to make make it more difficult for point-grabbers to receive KudoZ points. I believe that one could base a whole PhD thesis on this subject. Anyway, the problem I see with all these suggestions is that we are looking at it from the wrong end. We are all assuming that ... See more There is always some discussion on the KudoZ system and how to revise it or make it better. I cannot remember one single discussion where people would not start debating about the pros and cons of this and that or how to make make it more difficult for point-grabbers to receive KudoZ points. I believe that one could base a whole PhD thesis on this subject. Anyway, the problem I see with all these suggestions is that we are looking at it from the wrong end. We are all assuming that askers will review *all* criteria (including time of posting, confidence level, references, etc.) before choosing an answer. I believe that this is not the case most of the time. I rarely look at the time the answerer posted his solution though I do look at the confidence level and references. However, in the end, what really matters, is the answer given. Most (serious) translators already have something in mind when posting a question (unless they really can't make any sense of their source text). Consequently, they will look out for something they've been "looking for", something that "makes sense". Answerers who are not seriously interested in mutual help and just go for the points will find a loophole in almost any system, no matter how often we revise it or how good we try to make it. If we really want to make the KudoZ system better we have to take a closer look at the point-grabbers themselves. Why are these people trying to grab points? Exactly, they are after the points. This is as illogic as buying something on ebay that you can get cheaper at a local shop, but hey, people don't want to buy something, they want to *win* it. And that's exactly why point-grabbers are trying to *win* points. Of course, the side effect of a higher ranking in the directory is welcome, but - let's face it - that's just a side effect. I for one believe that we have to face the truths here: either abolish the point system or live with the point-grabbers side-by-side. Sonja ... and I know, that is not what you wanted to hear.
[Edited at 2005-05-16 07:24] ▲ Collapse | |
|
|
Mats Wiman Sweden Local time: 10:53 Member (2000) German to Swedish + ... In memoriam Beside my point Mike. | May 16, 2005 |
Mikhail Kropotov wrote: The asker is likely to benefit from more thorough answers Yes - IF - s/he is not to deliver in 37 minutes. Mats | | |
GILOU France Local time: 10:53 English to French Quick answer | May 16, 2005 |
I think it's a good thing that Kudoz should be fast. As a fast answerer, I'd like to say that the first answer is a proposal which serves as a guidance to the following answers. It's a question to educate the asker and to teach him to be patient before choosing. I happen to become an asker and it's very unpleasant to have a Trados segment open and to wait for 15 minutes before an answer occurs. Even If I get a rough guess as an answer, it allows me to fill the gap and proceed with my translation... See more I think it's a good thing that Kudoz should be fast. As a fast answerer, I'd like to say that the first answer is a proposal which serves as a guidance to the following answers. It's a question to educate the asker and to teach him to be patient before choosing. I happen to become an asker and it's very unpleasant to have a Trados segment open and to wait for 15 minutes before an answer occurs. Even If I get a rough guess as an answer, it allows me to fill the gap and proceed with my translation even if I come back to my segment to edit with an other answer. All of us tend to be angry when we can't open the GDT or Eurodicautom. Kudoz gives the possibility to get fast answers so let's enjoy our system and forget the matters of point grabbing because when it comes to difficult sentences, the 1st sentence is rarely chosen. And a translator can only reply quickly when it is his specialization fields.... ▲ Collapse | | |
Jaroslaw Michalak Poland Local time: 10:53 Member (2004) English to Polish SITE LOCALIZER If only it could be so... | May 16, 2005 |
GILLES MEUNIER wrote: And a translator can only reply quickly when it is his specialization fields.... Unfortunately, it is not true... That is why we have this discussion in the first place: because many "quick" answers are of poor quality. I agree that the original suggestion is flawed: people might be stealing other answers. It sometimes happens even now. I would change it a little bit: hide all the answers for ten minutes after the question is asked. When ten minutes passed, they would all be seen at once, so they would be recognized by quality (and could not be stolen). Of course, this would require the Asker to wait for ten minutes for the answer, so the change might not be popular. The idea of "urgent" questions is interesting, too. If properly implemented and described ("Warning! The answer to this question might not be perfect!" ) this might work. Or, even better, introduce "no hurry" questions. This would indicate that the Asker is in no rush and requires answers that are well explained. Relevant examples and references for those questions should be obligatory. The closing period would be longer, maybe three days, maybe a week. And the questions might be displayed with a proper filter, so that it is possible to go back to questions that were asked some time ago. | | |
Pages in topic: [1 2 3] > |