2

Hello friends at Gengo,

I have asked Gengo for an audio alert to be sent to my computer when new jobs are posted. 

Reasons:

1. I can leave my desk or the Gengo tab, and not have to carry my phone or laptop around the house, but still be alerted to jobs when they hit the job queue on the jobs page, long before job notices are sent out by RSS/email (when it is too late). I can install speakers in the yard, even.

2. Leaving my desk means that I can do other kinds of work, exercise, do housework, cook, or spend time with my family, without worrying that I will miss good jobs. All good things for life-work balance.

Status:

Gengo's response to this request has been to put it in the requests queue, but since I am the only one who requested it, the priority is low.

To the Gengo translators:

Please second this request if you agree that an audio alert would also help you.

Or if anyone has good alternatives, please share.

Best to all,

Betsy

 

 

2 comments

  • 0
    Avatar
    mirko

    Problem is that in several pairs being notified of a new job is such a rare occurrence that you'd risk having a heart attack if you happened to hear such an unknown and alien sound, and, should you survive that, by the time you left what you were doing and got to your PC, the job would've already been gone for ages (and maybe have already been submitted). Besides, I think you'd be better off with a RSS reader that already supports such a feature. Since what Gengo does is send email notifications every 5 minutes (at the shortest) and update an RSS feed based on the last time you checked it (but without having an RSS reader per se), I'm not sure how they would even go about doing what you suggest on their side, playing a sound on your end...

  • 0
    Avatar
    Alexander

    Your idea sounds good (no pun intended) but also has a drawback. I think part of the success of RumpelstiltskinRSS (200+ translators gave it a try, some 50 use it daily) is based on its audio alert. Another part is that it is optimized for a quick attempt to grab the job once the user has rushed back to the device.

    For optimal chances to snatch a job, you have to be aware that it is available in the first place. Therefore I agree it would be great if Gengo implemented an audio alert in their interface in the spirit of RumpelstiltskinRSS. However, in your proposal every translator in your pair would be alerted at the same time: the very moment the job arrives. This would only add to the heat of the rat race, and systematically disadvantage those with the slowest internet connection.

    In the current approach, with a cached RSS feed, there is some delay between the arrival of the job and you being notified about it. While this seems to be a disadvantage, it is actually beneficial. All translators are equally affected by the delay, so the negative effects more or less cancel. Even beter, the delay is more or less random, because the individual RSS readers of all translators are not synchronized. Hence there is some spread in the time translators are notified about any job.

    It may happen a job becomes available and is taken in between two times your reader checks the feed. In that case, you will miss out on it, but that won't spoil your day because you are not aware the job ever existed. Similar, if you do see a job, most likely some other translators don't see it yet in their reader, improving your chances to grab it thanks to reduced competition. So we should cherish the current caching mechanism, and in some pairs even beg Gengo to increase the time after which the cache expires (currently 1 minute) for an even better dilution of the crowd who gets to see (or hear, in your proposal) any specific job.

    Generalizing along these lines, the best approach would be to make any job available to only a small subset of the translators, picked at random, such that on average the pick up time will still be short but also the number of disappointed translators will be low. This way, it should be possible to find a happy medium between the conflicting interests of Gengo (short pick up time) and the translators (relaxed job taking).

    In fact, some time ago I sent a detailed proposal to Gengo for an algorithm that would do precisely this. It is a very general algorithm, in that with appropriate setting of the parameters it boils down to either direct allocation or the current rat race as special cases. The happy medium would be reached by choosing the parameters somewhere between the extremes. The algorithm is also future proof, in that it includes hooks to increase or reduce chances per translator in a controlled way, like better chances for translators who deliver better quality, lower chances for translators who let jobs expire (slowing down the system), etc. It could even take translators' preferences into account, like the time of day they are available and the minimum or maximum job size they want, should Gengo provide an interface where translators can supply this information to the system.

    Gengo showed interest, but it never worked out due to a triviality. (They wanted to Skype with me about it. I don't have Skype, and I procrastinated wading through Skype's terms and conditions.) I hereby invite Gengo to resume the discussion with me via email about the algorithm I proposed some time ago.

Please sign in to leave a comment.