Although Gengo has announced that it had been improving its job notification system, it's still a big problem that it's really hard to get a job with some other translators who obviously use a fradulent method to get notified about newly posted jobs because it's literally impossible to get one! I mean, I get notified by e-mail about a new job and only minutes, sometimes seconds after I get notified, I click on the link to go to the job and it says it's been taken! This is so frustrating! IF GENGO WANTS TO EXPAND AND BE BETTER COMPANY, IT IS HIGH TIME IT STARTED CONSIDERING THE TRANSLATORS' SATISFACTION FIRST! Thank you!
50 comments
You see Erika! You're even lucky grabbing some jobs! I am not even seeing jobs at all! My RSS is set at 32 seconds yet still nothing. Maybe there are no jobs at all for the Fre>Eng language pair, or some sort of mafia is going on. I reiterate my call for Gengo's administration to step in immediately. This attitude, if founded, doesn't reflect the image of Gengo. Something should be done about collection allocation, and immediately. Translators' satisfaction, not only that of customers, should be taken into consideration. There can't be customers without translators and vice-versa. The overall state of affairs is so discouraging, but we're fighters...
@Hilarious A. I work the French-English pair and am laughing that they actually sent you an e-mail saying that they need translators in this pair and that you'd start working right away. Aside from one big client for whom you must be a preferred translator, there are nowhere near enough jobs to keep the translators who have been already working in this pair for years busy. And even for that client, it's a fight between us preferred translators to get jobs. The only way to get any jobs at all in this pair is to camp out at the dashboard and refresh. If you aren't prepared to do that, then forget it. Forget about e-mails and signing in. If you have to sign it to your account, the game is over. You have to be there, signed in, and waiting and then be the lucky one to click faster than everyone else.
As for everyone's demand that Gengo change the way jobs are allocated, well, all I have to say is that we all joined up knowing this is the way Gengo works. I find it a bit laughable that people are demanding this be changed to be honest. If you want to bid on jobs, there are many other websites that will allow you to do that. If you want jobs allocated to you, then there are places you can go for that as well. Why should Gengo change its whole business model because some translators are unhappy that they can't get jobs?
@LB - In all fairness, Gengo has never really addressed the "bot" issue. They just wrote their use is forbidden in some remote corner of the support forum, but, as far as I remember, no Gengo representative has ever explained how they are policing and enforcing that rule (actually, if I remember correctly, they once wrote somewhere they can't...), how many users have been "caught in the act", what measures were taken against them and to counteract such practices, etc.
I believe these are quite legitimate concerns that shouldn't be so hastily dismissed, belittled or laughed at.
Hi Mirko. No, I agree about the bot issue. My post was in no way referring to that. The only thing I'm laughing at is the absurdity of hiring more translators where the last thing is needed is more translators because there already isn't enough work to go around. I believe it's the same problem in your language pair.
As for bots, I agree that some more transparency on that issue would be nice. Personally, I have noticed suspicious bot-like activity in the past and reported it along with the translator ID number. I don't know what Gengo did with that information, but that suspicious activity pretty much immediately ceased after it was reported so I believe they did do something about it.
@LBTranslations. Thanks a lot for this valuable contribution. It's very honest and straightforward. I'm also lucky to work on a website where bidding is the only way out, and from time to time I'm hired by clients, and I'm also hired by a company where jobs are directly allocated to me. But you could spend up to two weeks at times before you're allocated projects, and there are times when you work non-stop. I just thought I could use Gengo to fill in my downtime with this company. But the way I see things now, getting a Fre>Eng job on Gengo is like finding a needle in a haystack.
Hi everyone,
instead of just grumbling here, let's draft some questions we could address to Gengo. Personally I have to say that in my pair (EN->DE) the workload is okay even though I am not receiving jobs as a preferred translator. I can pick up several jobs a day by using a standard RSS notifier and by looking into the dashboard now and again. So it obviously has also to do with which language pair you are working. Of course, once I see a job, I have to grab it as fast as I can without taking the slightest look at it. And of course it often happens that someone else was faster. But I do not attribute this to bots or other forms of cheating, it seems to me this is just the normal result of who clicks faster. I still think there could be better ways of allocating jobs that would not force me to be present at my computer all the time with the finger hovering over the mouse button.
My questions to Gengo would be:
- How is Gengo going to address translator frustration regarding job allocation and (resulting) low workload?
- What does Gengo do to prevent unfair job grabbing methods like bots?
- What is the threshold at which Gengo would stop recruiting more translators for a language pair?
@Megan, maybe you could give us some insights. TIA.
@LB - Thanks for the clarification. Sorry, guess I had partially misunderstood your previous post. And yes, the seeming(?) absurdity of keeping tests open in already overcrowded pairs is something that has been discussed at length here, for years, but I'm not sure we've ever received some kind of reply from Gengo... have we? The only thing which is safe to assume, I believe, is that by coming up with specific rules concerning bots (and refresh rates), Gengo itself has implicitly acknowledged that this problem does exist, although, as I was saying, we were never told how it is being dealt with (or if it has been dealt with and solved).
And indeed, the situation in my pair is definitely bad (as in *BAD*) too, so much so that in most cases, when I'm able to grab a job, it's because I have been selected as PT (probably together with only a few others). I took a look at the "open data" and it seems the "active translators" are now 15 (roughly 23%) more than a few months ago, but the workload didn't certainly follow suit...
@kvstegemann - Good questions. Seconded.
@Hilarious - Personally I must say I do hate "reverse bidding" and think it would be a bane in an anonymous and crowdsourced system like Gengo. To be more specific, (the lowest) price would become the only deciding factor, as all translators are the same in the eyes of the client here on Gengo (just a number). The client wouldn't be able to choose Translator X, with 10 years of proven experience in IT, marketing, patent translations, ATA certification, MA in Translation Studies offering their services at 100 and Translator Y, with no experience, no accreditation, nor formal education, offering the "same" services at 5... IMO, that would be like jumping out of the frying pan into the fire.
I just saw an official Gengo post on my Facebook wall stating that they were opening testing in a number of pairs (including mine!).
All pairs now open are *to* UK English – and they are from: Italian, German, Dutch, (European) Spanish, Danish, Swedish, and Polish.
I had a pretty good month (by far the best month I've had since joining Gengo a year ago), easily getting tons of work, and was enjoying it because I knew it was probably temporary.
Looking back, the # of words I translated this past month is about 60 to 70% of the total number for the PRIOR 11 MONTHS combined.
It was fun while it lasted, and I sure learned a lot!
@kvstegemann On the topic of questions to ask, although not strictly related to the amount of work available, I'd like to know why some of the texts used in the various pre/standard/pro tests contain mistakes and why, despite notifying Gengo about it, they aren't corrected. How can someone be marked on a translation done from a source text with mistakes is something I really cannot understand.
@Cecilia B. - I don't see a problem here. In real world texts, mistakes do occur. A translator should be able to handle that (as long of the intended meaning is clear).
as long as
@Alexander I'm very well aware that errors occur. Unfortunately, unlike normal translations, you cannot leave comments to the reviewer of your standard/pro/ultra test.
Say that I see that word X is wrong, for example because it is a colloquial word whereas the rest of the text is very formal. If I translate it with an equivalent colloquial word I "carry on" the mistake found the source text. If I translate it with an equivalent formal word, the reviewer has all the right to say I made a mistake.
A friend of mine recently counted at least 20 errors in her standard test, in my (and her) opinion 20 errors in a 300-long text is a bit too much. I'm not talking about a simple misplaced comma or a missing period mark. I'm talking about errors like unnatural collocations, big punctuation mistakes…these are serious errors that shouldn't be there and that, without a doubt, can make it hard for a non-native speaker (since I'm referring to the source text) to understand what is being said.
The only tests where errors are, and should be, accepted are revision tests because the aim of those tests is to measure how good you are at spotting mistakes and correcting them. Normal translation tests shouldn't be filled with mistakes.
@Cecilia: A real problem can occur here only if the initial multiple-choice test is flawed, because in this case the interested translator would be rejected from the start and would have no chance to "appeal". And if you make a statement like this, you should name the language pairs where this problem occurs.
Apart from that, it seems to be a fact of the daily life of a translator that source texts contain mistakes, up to the point where they are outright crappy. So you could call this just an exercise ... :)
@kvstegemann I don't see the point in mentioning the language pairs, since there are many questions available and each translator gets a different selection. Anyway, I saw one case in the German > English pre-test, and I think my friend was referring to the recently launched Italian > British English test.
Again, in a real-life situation you get a text with mistakes AND you can notify the customer/reviewer about it. When you take the test you can communicate with the revisor only after you have taken the test and, I think, only if you don't pass it. I wouldn't say it's the same thing.
I'm with Cecilia on this. If the test is for proofreading, then OK, the errors in the translation are an integral part of the test (and even then the error should be clearly identifiable as such), but not in a translation test, which should only be testing your abilities in translating a text from source language X into target language Y, also seeing how in this case you're not even allowed to ask for clarifications, leave notes, etc. Besides, I don't think the style guides mention how errors (or possible errors) in the source should be handled, do they?
A translation test is totally different from the average everyday translation you do in standard working conditions, where yes, you're bound to find errors, unclear passages, weird wording, etc. in the source, in which case you often need to ask the client about some of those in order to confirm they actually are errors, clarify the intended meaning of the affected passages, and to even know whether you have to "correct" them or not (and how)...
It sounds kind of absurd to me that the source text for a test (about 250 words, on average) should contain something like 20 errors (based on what Cecilia wrote). One error every 12 words? Really?
@mirko Thanks a lot, I feared I hadn't been clear enough. I did several translation tests in the past (off Gengo) and it never happened to see a source text with mistakes, unless, as I said before, it was a proofreading test. Also, as you correctly pointed out, we don't have an "error evaluation table". If all the answers are wrong I can certainly think that a spelling mistake or the use of a false friend is more serious than a sentence without the closing period mark, but that's my personal judgement.
Just be clear, I also took several tests on Gengo and aside from that once sentence in the German > English test I don't recall any major problem so if these errors affect only certain pre-test answers and some standard/pro tests texts they could be easily fixed.
Unfortunately my friend's internet connection isn't working well today so she cannot join us. Since we cannot talk about the test contents, she didn't make any specific example. She only mentioned she saw quite a lot of errors (most we minor, but still, they were errors) and was a bit worried that her test couldn't be evaluated properly if the source text had mistakes.
Super quick follow-up (sorry for the double post, I wasn't able to edit it), my friend is still with no working internet connection, but she sent me via an update via SMS. She wrote down an exact list of all the errors and the final count is 18. Errors types are: unnatural sentence constructions, unclear passages, wrong punctuation, repetitions. She contacted Gengo support yesterday, but although they will look into the pre-test mistakes, they won't do anything for the standard test ones. They told her to make the translation anyway and request a revision should she fail the test.
(Sorry in advance from eventual typos, I'm writing from my mobile phone).
I really don't know if there's "job thieves" involved, but there's something really weird happening with available jobs and the RSS (at least for EN-ES)!
Usually, all jobs would pop up on my RSS (set to 1 min), even if they're taken. However, at random times I've updated the "available jobs"page and actually found available jobs listed that never make it to the RSS. Of course, they are taken by the time I click on the job link.
Do you know why is it that some jobs don't make it to the RSS feed?
Thank you.
Hi indiralena, I have the same thing in the en-rus pair.
Hello. I have addressed this problem before with support, but nothing has been done really. There seem to be job thieves in the English > Arabic language pair. They just said that other translators are picking it up faster than I do. I'm literally camping on the platform and every time there is a big job, once I open it, it's a blank page and there is not even the "collection not available" notification. It's really weird. This has been going for over 3 months now and I fell like I'm wasting my time. I hope there is a solution to this problem or even a deeper investigation.