Is there a good reason for keeping the confidence level indication? Thread poster: writeaway
|
|---|
I realise the confidence level issue has been discussed at length before, but it seems to have evolved in such a way that many people, site newcomers in particular but not only them, now use it as a 'hard sell' of their answer. So ultimately it has become totally meaningless. People use level 5, provide no refs and concoct some answer trying to get "points" (which are not up for discussion here). Plus a lot of newcomers seem to be using the site as a substitute for own knowledge and are very eas... See more I realise the confidence level issue has been discussed at length before, but it seems to have evolved in such a way that many people, site newcomers in particular but not only them, now use it as a 'hard sell' of their answer. So ultimately it has become totally meaningless. People use level 5, provide no refs and concoct some answer trying to get "points" (which are not up for discussion here). Plus a lot of newcomers seem to be using the site as a substitute for own knowledge and are very easily misled by exaggerated confidence levels. The combination of high confidence and high degree of error now occurs far too often.
I feel that if questions were to be judged on their merits and not on the self-awarded confidence level, people would be forced to think for themselves or pay more attention to professional peer comments. ▲ Collapse
| | | | | Is there a reason for keeping the confidence level indication? | Feb 21, 2005 |
Good point, writeaway. It is sad to see such practices in KudoZ. I don't think the confidence level should be abandoned though. It is part of an overall strategy to help askers (who are usually getting paid for their work) decide how much value to place on a proposed translation. I think your posting is a step in the right direction: when things aren't working as intended we need to talk about it. Some of the abusers might even take notice.
---
confidence man
n.
A... See more Good point, writeaway. It is sad to see such practices in KudoZ. I don't think the confidence level should be abandoned though. It is part of an overall strategy to help askers (who are usually getting paid for their work) decide how much value to place on a proposed translation. I think your posting is a step in the right direction: when things aren't working as intended we need to talk about it. Some of the abusers might even take notice.
---
confidence man
n.
A man who swindles his victims by using a confidence game.
Confidence trick
A confidence trick, confidence game, or con for short, (also known as a scam) is an attempt to intentionally mislead a person or persons (known as the mark) usually with the goal of financial or other gain. The confidence trickster, con man, scam artist or con artist often works with an accomplice called the shill, who tries to encourage the mark by pretending to believe the trickster .... in more general use, the term con is used for any fraud in which the victim is tricked into losing money by false promises of gain.
http://en.wikipedia.org/wiki/Confidence_trick
[Edited at 2005-02-21 00:54] ▲ Collapse
| | | | writeaway 
French to English
+ ...
TOPIC STARTER | It's harder to 'bluff' if you can't show how confident you are | Feb 21, 2005 |
[quote]Kim Metzger wrote:
"... I don't think the confidence level should be abandoned though. It is part of an overall strategy to help askers (who are usually getting paid for their work) decide how much value to place on a proposed translation."
My point is that it is not really an indication of anything since so many people use it as a ploy to convince others that they are offering "the" solution. Nothing at all stops people from rating themselves as they wish. And since there are now so many new people who clearly are using Proz as a work tool, not as a site for participating in the translation community (just look at the questions asked compared to questions answered numbers to get some idea of involvement), those Askers are not around the site long enough to see who is really reliable and who isn't. So they seem to rely heavily on the confidence level which is basically a bogus indicator. Does it really matter what a translator thinks about his/her answer? Imho, peer comments are much more valuable and when you see a 5 confidence level and the peer comments are all elsewhere, then one should draw the right conclusions. But there are now many Askers who really have no idea about the answer themselves and often seem very swayed by high confidence level with accompanying 'attitude'. It's starting to look like a game of bluff and if there were no confidence level shown, the game would end.
| | | | Subhamay Ray (X)
Local time: 19:30
English to Bengali
+ ...
| Extremely perceptive | Feb 21, 2005 |
Writeaway wrote: And since there are now so many new people who clearly are using ProZ.com as a work tool, not as a site for participating in the translation community (just look at the questions asked compared to questions answered numbers to get some idea of involvement),
Nothing can be closer to the truth. I would judge a translator by the questio... See moreWriteaway wrote: And since there are now so many new people who clearly are using ProZ.com as a work tool, not as a site for participating in the translation community (just look at the questions asked compared to questions answered numbers to get some idea of involvement),
Nothing can be closer to the truth. I would judge a translator by the question he/she asks and not his/her answers. There are translators, yes many so-called 'leaders', who never ask questions and apparently feel impatient to answer as soon as a question pops up. I wonder if they don't encounter any problem while translating -- no obscure phrases, no difficulty in finding the right terminology. Some people indeed are like gods!
It would be an interesting statistics to find out how many questions a translator has asked and the quality of the questions. That would be a better indication of a translator's worth than his/her answers.
I also agree wholeheartedly that the confidence level indication has become ridiculous. The only motto now is to amass Kudoz points which obviously helps in getting jobs from this site.
Thanks for this thread. ▲ Collapse
| | |
|
|
|
Kirill Semenov
Ukraine
Local time: 17:00
Member (2004)
English to Russian
+ ...
| A possible purpose of the confidence indicators | Feb 21, 2005 |
Subhamay Ray wrote:
It would be an interesting statistics to find out how many questions a translator has asked and the quality of the questions. That would be a better indication of a translator's worth than his/her answers.
First of all, I totally agree that the confidence indicator as it is now is often useless or misleading. When a subjective number (an answerer's own confidence) is added up to some objective number (the number of agrees) we get a strange result which really shows nothing.
But I believe there were some reasons why the proZ staff implemented this indicator. It plays it's part in the formula for calculating kudoz reliabilty ratio which was developed a long time ago and is still described in the kudoZ FAQ (but never implemented in practice). I cannot think of any other reason why kudoZ zone needs confidence level. Moreover, I find it a bit annoying to select it again and again, since this is a mandatory field of kudoZ answers.
If the site is to keep the indicator, it should be used more effectively -- say, it may be given more "weight" for askers, but with penaltizing answerers who tend to be over-confident (for instance, a `5-confident' answer which gets a couple of disagrees, is, most likely, rather a _self_-confident one).
However, my personal opinion is that the confidence indicator well may be altogether discarded. I doubt that kudoZ system would somehow change without it. At least, we would not feel *furious* when seeing yet another wild guess veiled by confidence 5.
[Edited at 2005-02-21 07:05]
| | | | | Abuse has made the indicator meaningless | Feb 21, 2005 |
In the French - English pair, I have definitely noticed this. Certain answerers systematically use "5" as a way to sell their answers. This, combined with snide personal comments to other answerers ("oh puhleeease" or "come on" etc.), notes in all caps, and other violations of netiquette and the Proz.com rules definitey defeat the whole purpose of Kudoz by "bullying" the asker into choosing a certain answer rather than letting the asker choose based on the merit of the actual answers proposed. I... See more In the French - English pair, I have definitely noticed this. Certain answerers systematically use "5" as a way to sell their answers. This, combined with snide personal comments to other answerers ("oh puhleeease" or "come on" etc.), notes in all caps, and other violations of netiquette and the Proz.com rules definitey defeat the whole purpose of Kudoz by "bullying" the asker into choosing a certain answer rather than letting the asker choose based on the merit of the actual answers proposed. I think that new or infrequent users of the tool are the ones who are particularly affected by this.
As writeaway mentioned, these types of answerers rarely provide references or explanations, other than "that's just it" or "this is THE answer" or "because I just know" or "one just says it this way in English".
It is true that regulars on the site are able to judge the credibility of answerers based on this kind of behavior, but new or sporadic users don't have that advantage, which is a shame for them and for the resulting Kudoz glossary.
Thanks for raising this issue, writeaway.
Regards,
Sara ▲ Collapse
| | | | Javier Herrera (X)
Spanish
Subhamay Ray wrote:
It would be an interesting statistics to find out how many questions a translator has asked and the quality of the questions. That would be a better indication of a translator's worth than his/her answers.
You can see all questions asked by anybody on their profile. The quality is up to you to judge.
| | | | CMJ_Trans (X)
Local time: 16:00
French to English
+ ...
| I agree ... but... | Feb 21, 2005 |
How do you decide what confidence level to give yourself? You might as well ask how long is a piece of string...
I, for one, have to admit that I never know what to do for the best. Even when I'm sure, I hate putting '5' because it reeks of overconfidence: nobody's perfect, after all.
By scrapping this feature it would be possible, indeed, to weed out those who use it to force their views through at all costs. At the same time, there should be some way of indicating to ... See more How do you decide what confidence level to give yourself? You might as well ask how long is a piece of string...
I, for one, have to admit that I never know what to do for the best. Even when I'm sure, I hate putting '5' because it reeks of overconfidence: nobody's perfect, after all.
By scrapping this feature it would be possible, indeed, to weed out those who use it to force their views through at all costs. At the same time, there should be some way of indicating to "askers" that one is more or less sure of what one is saying. I am sure that I am not the only person who sometimes offers comments to try to direct others along the right track, even if I don't have the perfect answer myself. I usually say in so many words what I am doing but I'm sure others have neither the time nor the inclination to do the same.
As devil's advocate I would say that the danger of scrapping "confidence" level is that the only yardstick would then be peer comments. Don't get me wrong: I've got nothing against peer comments but, while we are on the subject of pushy tactics, there are cliques of people who stick together like glue and vote for each other (there are also cases of people who vote systematically AGAINST other people) so even peer comments are neither reliable nor disinterested. There are people whose questions I now ignore, even when I know the answer, simply because I've seen that they systematically choose answers by the same people (even if these answers are not always correct) and will always ignore mine. The old expression including the words "without butter" comes to my mind where such people are concerned.
There is also the problem that if you give your answer hours after the original exchange (due to time zones), the peer comments have already been entered and new ("late") answers will be unlikely to get much support, however excellent they may be.
So, despite Writeaway's comment about this not being about points, we cannot really ignore the fact that the problem will always be with us in some form or other for as long as the points system applies. Sorry to come back to this but "Facts is facts!" as they say. ▲ Collapse
| | |
|
|
|
writeaway
French to English
+ ...
TOPIC STARTER | you can see the questions answered, not the questions asked | Feb 21, 2005 |
Javier Herrera wrote: Subhamay Ray wrote:
It would be an interesting statistics to find out how many questions a translator has asked and the quality of the questions. That would be a better indication of a translator's worth than his/her answers.
You can see all questions asked by anybody on their profile. The quality is up to you to judge.
On the Proz profile page, people can show Kudoz points and there one can see the questions they answered, not asked.
The only chance to see questions asked is when a registered member asks a question and still has questions open. then one can click to see all questions. But this is not a feature on the profile page. And anyway, that't not the point.
| | | | James Calder
United Kingdom
Local time: 15:00
Spanish to English
+ ...
| KudoZ needs an overhaul | Feb 21, 2005 |
I agree entirely with you, writeaway. There are several changes that need to be made to KudoZ (question/point averages among them, but we won't go into that here) and doing away with the confidence level indicator is one of them. As an occasional asker I never look at the confidence level. It's says absolutely nothing to me about the quality of the answer, and appears to be used as a substitute for providing an explanation and/or references. I'd rather have a level 3 answer with an explanation t... See more I agree entirely with you, writeaway. There are several changes that need to be made to KudoZ (question/point averages among them, but we won't go into that here) and doing away with the confidence level indicator is one of them. As an occasional asker I never look at the confidence level. It's says absolutely nothing to me about the quality of the answer, and appears to be used as a substitute for providing an explanation and/or references. I'd rather have a level 3 answer with an explanation than a level 5 one with the ubiquitous "That's it" or "xxxx" or the eternally irritating "HTH" on its own (it doesn't help, actually).
By the way, we seem to keep having recurring discussions about the merits and drawbacks of KudoZ. I'd like to ask Henry if there are any plans afoot to change the way it is currently set up.
Regards
James ▲ Collapse
| | | | Lia Fail (X)
Spain
Local time: 16:00
Spanish to English
+ ...
| subjectivity | Feb 21, 2005 |
[quote]CMJ_Trans wrote:
At the same time, there should be some way of indicating to "askers" that one is more or less sure of what one is saying.
quote]
It's SO TOTALLY AND ABSOLUTELY subjective! Consider all the subjective input that goes into a confidence rating of 0-5:
a. personal arrogance/over-confidence
c. professional arrogance/over-confidence
b. a ploy purely to try and win points
c. ignorance, lack of experience
d. a why not attitude, it's so subjective anyway
etc.
Fundamentally, it's down to the psychological make-up of the answerer: are they more or less pushy, more or less self-confident, greater or lesser risk-takers, more or less opinionated, etc?
What's more, it's the asker who has the context, which is worth far more than any confidence index + agrees from people who don't have that context. It's also the asker who will/learn to know which answerers are reliable (usually from more modest self-raters anyway, becuase an element of experience in translation is knowing what you DON'T or MAY NOT know or acknowledging that without the context of the text, you can't be sure).
As Sara pointed out, it's actually an incitement to irritation by site users who see evidently erroneous answers accompanied by a rating of 5, which just knocks the credibility of the site on its head. I have never made a comment, but I have departed a question in disgust on seeeing some bad answers + high confidence levels, and especially when they are the chosen answer.
Almost everything about translation is subjective, so you simply can't scale things, but most of all, not confidence in answers. It would be far better to have answer boxes that included an obligatory box for justifying one's answer in a sentence or two, with supporting material from the WWW, and where this wasn't available, another kind of justification - qualitative, not quantitative.
| | | | Nesrin
United Kingdom
Local time: 15:00
English to Arabic
+ ...
| Keep the confidence level | Feb 21, 2005 |
Personally, I like to make proper use of the confidence level. I don't shy away from the "5", because I am sometimes absolutely confident that my answer is correct (e.g. Q: What does the German word "Tisch" mean? A: Table. Confidence level: 5) Other times, I want to make sure the asker knows that I am merely guessing, or that I think my suggestion is pretty good, but there may be better.
Don't let the people who abuse the system spoil it for the rest of us. Sooner or later, they ar... See more Personally, I like to make proper use of the confidence level. I don't shy away from the "5", because I am sometimes absolutely confident that my answer is correct (e.g. Q: What does the German word "Tisch" mean? A: Table. Confidence level: 5) Other times, I want to make sure the asker knows that I am merely guessing, or that I think my suggestion is pretty good, but there may be better.
Don't let the people who abuse the system spoil it for the rest of us. Sooner or later, they are exposed as such, and hopefully the "agrees" other answers will get will push these bogus "5s" further down the ranking. ▲ Collapse
| | |
|
|
|
Javier Herrera (X)
Spanish
writeaway wrote: Javier Herrera wrote: Subhamay Ray wrote:
It would be an interesting statistics to find out how many questions a translator has asked and the quality of the questions. That would be a better indication of a translator's worth than his/her answers.
You can see all questions asked by anybody on their profile. The quality is up to you to judge. On the Proz profile page, people can show Kudoz points and there one can see the questions they answered, not asked. The only chance to see questions asked is when a registered member asks a question and still has questions open. then one can click to see all questions. But this is not a feature on the profile page. And anyway, that't not the point.
On somebody's profile page, go to the Kudoz tab, then there ara another two, one for asked questions, one for answered questions.
| | | | Tony M
France
Local time: 16:00
Member
French to English
+ ...
SITE LOCALIZER | Keep the confidence rating - I agree | Apr 25, 2005 |
Nesrin wrote: Personally, I like to make proper use of the confidence level. I don't shy away from the "5", because I am sometimes absolutely confident that my answer is correct...
Don't let the people who abuse the system spoil it for the rest of us. Sooner or later, they are exposed as such, and hopefully the "agrees" other answers will get will push these bogus "5s" further down the ranking.
I couldn't agree more! I think it is very important to be able to give Askers an indication of one's own faith in an answer, and like Nesrin, I rarely use '5', but when I do, it's because I'm really sure (always, of course, with the caveat than anyone may turn out to be wrong with hindsight or more context!)
I'd even like to be able to edit my confidence ratings, so as to look less stupid without actually having to hide my answer, or sometimes, to boost the confidence level in the light of peer support.
I use it much more as a way of saying "here's my idea, but I'm not sure..." than anything else.
I think we are most of us reasonable and try to be objective, and I too don't think we should spoil a still-useful system just for that minority who abuse it. In any case, regular users soon get to know who to trust or not...
A couple of suggestions: as has been pointed out, it is often newcomers who perhaps unwisely give high ratings; so could all new members be given some guidelines, perhaps the first 20 times they anwser a question or something --- I know I found privately-sent moderator feedback most helpful when I was a novice too, and had not received any guidance on the matter.
Likewise, concerns have been expressed over occasional users setting too much store by the confidence rating; well, 'caveat Quaestor' --- those who join in regularly to the community will soon suss out the ones to be wary of, and anyone who is only an occasional user outgh to be exercising caution anyway. But it would do no harm to post a warning, say for non-member Askers, and for the first few times a new member asks questions, to the effect that confidence ratings should be taken with a pinch of salt.
I don't like the inference of one Poster in this Forum, who suggested that those people who answer lots more questions than they ask are into self-aggrandizement. I am probably one of those people, but that's certainly not my reason for doing it: I am just so acutely aware of not wanting to waste my colleagues' time, I go to huge lengths to research things privately, and really only turn to a KudoZ question as truly a last resort (and golly, how often has it been a life-saver!). On the contrary, those who answer more than they ask may simply be people who are good at research, which may make their answers even more authoritative. No, I really think we should be careful about making such blanket judgements about peoples' motives for answering or asking... there can be so many reasons...
[Edited at 2005-04-25 11:25]
[Edited at 2005-04-25 11:27]
[Edited at 2005-04-25 11:29]
| | | | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Is there a good reason for keeping the confidence level indication? | 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 » |
| | 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! » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
