| Stranica u temi: < [1 2] | Proposal for change of ranking calculation system Postavljač teme: ------ (X)
|
|---|
Pablo Grosschmid 
Španjolska
Local time: 21:27
engleski na španjolski
+ ...
U sjećanje | answerers opting out of points | Oct 9, 2005 |
Kim Metzger wrote:
We would need a way for answerers to opt out of the points system.
Fully agree. So they can give additional information without their average being penalised.
| | | | Daniel Coria
Argentina
Local time: 17:27
španjolski na engleski
+ ...
| Ranking sorted by overall points or by reliability ratio/precentage | Oct 12, 2005 |
This may have probably been suggested before, but it is perfectly possible -technically speaking- to have a ranking list and to allow the user/agency/outsourcer to sort it by total points (be it 3-month, 12-month, or all-time), or by reliability ratio (again, 3-month, 12-month, or all-time) calculated by dividing the number of questions selected by the number of questions answered, and expressing it as a two-decimal percentage.
This, of course, has one potential flaw: a person who h... See more This may have probably been suggested before, but it is perfectly possible -technically speaking- to have a ranking list and to allow the user/agency/outsourcer to sort it by total points (be it 3-month, 12-month, or all-time), or by reliability ratio (again, 3-month, 12-month, or all-time) calculated by dividing the number of questions selected by the number of questions answered, and expressing it as a two-decimal percentage.
This, of course, has one potential flaw: a person who has answered only 4 questions and received points for 3 of them would show a reliability ratio of 75.00%, thus making a very "early" climb in the ranking; but than can easily be solved by allowing only those answerers who qualify into the list through their total points earned, as the current system does.
In that way, we would still have the current ranking system -the fairness of which might be arguable or not- but it would be supported by an accuracy/reliability index. The list would show both indicators, and the viewer can choose to have it sorted by one or the other.
As usual, "Hope it helps!"
Daniel ▲ Collapse
| | | | | competitive /collaborative - let's keep a balance | Oct 13, 2005 |
The main reason I dislike the suggestion of ranking by ratio is that I'm wary of "processed data" - I prefer to have as much raw data as possible and be able to make the processing (and assess significance and sense of provided data) myself. So I'd favour a system where total, subtotals and ratios are all easily available, and outsourcers can list freelancers by their own criteria. Maybe, with the help of a few words of advice.
Also, I think that we should try to keep a balance betw... See more The main reason I dislike the suggestion of ranking by ratio is that I'm wary of "processed data" - I prefer to have as much raw data as possible and be able to make the processing (and assess significance and sense of provided data) myself. So I'd favour a system where total, subtotals and ratios are all easily available, and outsourcers can list freelancers by their own criteria. Maybe, with the help of a few words of advice.
Also, I think that we should try to keep a balance between the competitive and collaborative aspects of the kudoz system. Let's remember that its main purpose is supposed to be helping colleagues, not showing off. The asker,after all, is not supposed to chose the best answer, but the most helpful. So, if kudoz points can be an indicator of a tranlator's ability, they cannot be, and are not meant to be, a general purpose ranking system. Support of the community, willingness to help and free time, for instance, also matter. In fact, I think that an outsourcer who selects freelancers solely, or mainly, by kudoz ranking is probably a fool (but, of course, reviewing kudoz history may be a very good idea). BTW, in my opinion, if we want to improve the usefulness of the kudoz system to outsourcers, the first area that needs improvement is the classification of the questions by field and specialization.
I' am concerned, in short, lest "improvements" in the ranking system may actually discourage contributions that are both useful and in keeping with the spirit of kudoz (as I see it). I have seen (and made, too) several contributions that others have been able to use to propose a better solution (or one, at least, that the asker deemed more useful); and this is as it should be - it is a fact that often the most suitable answer can only emerge by a collaborative "trial and error" process. If such contributions will be actually penalized, many will stop making them - and the very usefulness of kudoz will be greatly diminished.
Ranking should be, in my opinion, an added benefit of kudoz, not its main purpose. ▲ Collapse
| | | | ------ (X)
njemački na španjolski
+ ...
POKRETAČ TEME | Excellent idea | Oct 13, 2005 |
And IMO the default ranking should be the ranking by reliability. I really want to underline that the best technical translator on proz in the combination german
| | |
|
|
|
------ (X)
njemački na španjolski
+ ...
POKRETAČ TEME | Impossible to complete my sentence. Seems to be a bug | Oct 13, 2005 |
????????
[Edited at 2005-10-13 23:28]
| | | | | Stranica u temi: < [1 2] | To report site rules violations or get help, contact a site moderator: You can also contact site staff by submitting a support request » Proposal for change of ranking calculation system | Trados Studio 2022 Freelance | The leading translation software used by over 270,000 translators.
Designed with your feedback in mind, Trados Studio 2022 delivers an unrivalled, powerful desktop
and cloud solution, empowering you to work in the most efficient and cost-effective way.
More info » |
| | Pastey | Your smart companion app
Pastey is an innovative desktop application that bridges the gap between human expertise and artificial intelligence. With intuitive keyboard shortcuts, Pastey transforms your source text into AI-powered draft translations.
Find out more » |
|
| | | | X Sign in to your ProZ.com account... | | | | | |
Login to reply/comment 
