You are viewing a single comment's thread from:

RE: Talk to the CEO - Utopian Translations Category

in #utopian-io6 years ago (edited)

As I told you in our quick chat on Discord, I wasn't going to write something, because it might sound as heavy accusations, however I changed my mind, as it might be good in the end. I'm not going to name anyone, as it might have been random or someone is going through a lot (and even though it is not right, I understand that it might happen from time to time.) I have discussed my concerns with a CM, for what it's worth.

Communication

Our category lacks communication. Real communication. Except from the weekly reports, there is virtually nothing else. Yes, we may be writing reports with all our thoughts etc, however having constructive dialog -instead of sweeping our problems under the carpet- will have a positive outcome.

There may be things in motion in the background, but having updates (as the case has been in the past 10 days or so) is much better and helpful for all of us (we won't have to say the same things over and over again)

Category in general

Fact: Translations category is using a lot of VP (or at least, it was, before the change in the bot). Not even a single member of the category denies that. But there are a lot of ways this category can't work in the same way as (random name of category) "development". I'm saying this as I'm a contributor to both Translations (as a Translator/LM) and Development categories

A translation of a project happens once and it might not need more strings to be translated in the future.

Developing a project is a recurring work that may never end, but it doesn't mean that new strings will be added that might need translation.

And, as already stated by @scienceangel:

First thing that should be considered is limiting the number of people within each team, because if more languages will be added, and if every language can add as much people in their team as they want, this is the recipe for disaster.

There can be room only for the best from the best from the best.

It's simply not sustainable that some teams have more than 10 members, while some have 3 - 5.

Scoring and complaints

According to the statistics I'm gathering for the Greek team, our average score per contribution is over 70. And no, I'm not favoring anyone (at least not on purpose). Let's take one of my last contributions as an example (this one.)

How would you score it? (with the exception of "How would you rate the semantic accuracy of the translated text?" as you can't know my actual accuracy if you don't know Greek) It's not a rhetorical question, I'm actually waiting for replies saying their answers. I can handle constructive criticism (as far as I know at least 😂)

Thing is, there are complaints that the Greek team is scoring very high (without saying it's abuse on purpose or justified). And to be honest, while the specific contribution was actually of high difficulty, with the current questionnaire at least, I've asked @ruth-girl to mark it as average to lower my score by telling her it was actually (and I quote myself) not that difficult. And on the other hand, I've suggested to a translator of my team not to submit a contribution because their translation was very bad, and they shouldn't seek a reward, even after reviewing and correcting their work. But that's not my point and I'm not waiting to hear "Bravo" or "Congratulations!" for any of those points, it's just "αυτονόητο" as we say in Greece (meaning self-evident)

I can easily lower my score even more. By submitting crappy posts (on average I spend 20 minutes on each contribution post, and I would gladly stop doing that), crappier translations (it's actually very easy to do that). I can also remove points from the contributions I'm reviewing (which would be unfair for someone who invested a lot of time in researching and translating correctly) and there is one question on this:

Do we want quality or quantity? We can't have both, as we can't expect people to spend 3 hours to translate 1500 words, with next to perfect quality, another 30 minutes to create a great contribution post, only to get less than $50. The same amount of work in a translations office in Greece would get $100+, and in other countries even more. I'm not asking to have the same rate, it's impossible, but punishing translator X (who does a great job), just because translator Y (who didn't do as good of a job as X) didn't get an upvote (which won't happen, as at the current rate utopian is voting, all contributions are getting their incentives) is bad for everyone, as good translators will not show the same interest in the future, and even worse, they may abandon the category.

To end this section of my comment, I will gladly take penalties (i.e. lower the % of my upvotes for the next X contributions) if the score I've gotten in my contributions is higher than it should have been, and if I've favored anyone. With the information we have, I did what I thought best.

Autonomous category?

To conclude my wall of text, I've been thinking lately if it would be better to make the category autonomous, not related to the other categories. Yes, we might use all the VP utopian has, given the opportunity. So instead of causing issues to everyone else, wouldn't it be better if we had a different voting bot and community account? Having a seperate delegation on that would help all the categories (again, with the new voting logic there are no issues, but I'm saying this just for the heck of it).


Note #1: if you think an LM is favoring someone on his team, discuss it with one of the CMs or one of the DaVinci staff. I would love to know if you think I've favored someone on my team so hit me up on Discord to discuss this 😉 Communication is key to making everyone better.

Note #2: I'm not saying anything about the questionnaire, as we've already submitted our suggestions as a team in a post on our community account, and my comment is already too long. And I wasn't going to write something 😂

Sort:  

There's a lot to discuss here but the first point that is immediately evident for me is: we are giving less than what a translator will get with their normal job. This is not a normal job and it is not meant to be. If you were contributing to those open source projects and utopian didn't exist, you'd get nothing more than a "thanks" and a good feeling for having contributed to a good cause. Additionally you aren't getting fiat but STEEM which is highly volatile and today may be worth X and tomorrow Y, so it's unreasonable to compare it with a standard currency income. This is not to underestimate and I definitely think it's a mistake to sell the idea we should try to give as much as a professional job will pay. What you are getting is an incentive. Creating a separate account for just the translations category isn't a bad idea per se as I wish rewards will be more and more decentralized, but it's also a long term solution and a solution for which Utopian can only help with exposure, that to say, Utopian won't create an account specifically to support that, but you and others may do so.

If you were contributing to those open source projects and utopian didn't exist, you'd get nothing more than a "thanks" and a good feeling for having contributed to a good cause.

I agree with this statement, 100%, and I'm not trying to make a direct comparison -not in the way it came out-. And I've been telling this to other people in the category in the past, together with "upvotes are not guaranteed and we are basically volunteers with a form of payment" and it proved to be quite an unpopular opinion (with responses ranging from "I don't understand why I would do something like that for free" to "You should stop saying that, it's your own opinion").

we are giving less than what a translator will get with their normal job

I surely am not asking for a raise, just voicing some concerns from others. The current reward/incentive/call_it_what_you_may scheme is more than great, for translators at least, but the questionnaire as we have pointed is imbalanced (and we are running in circles 😂) and it is too punishing. From the LM/proofreader point of view, it could use a small bump but I'm personally fine with it as is.

Creating a separate account for just the translations category isn't a bad idea per se as I wish rewards will be more and more decentralized, but it's also a long term solution and a solution for which Utopian can only help with exposure, that to say, Utopian won't create an account specifically to support that, but you and others may do so.

There are already a few accounts, but they lack delegations (at least, sizeable ones). DaVinci has one, the Greek team has one, and both are upvoting translation contributions, however our effect is minimal -for now at least, and we hope in the future we'll all be able to offer more-

Creating a separate account for just the translations category isn't a bad idea per se as I wish rewards will be more and more decentralized, but it's also a long term solution and a solution for which Utopian can only help with exposure, that to say, Utopian won't create an account specifically to support that, but you and others may do so.

Are you talking about a separate Utopian's account to support this category solely? Don't forget that you can't split delegations and if you found a way to split your VP, it would mean a loss for everyone.

I am telling exactly the opposite. Utopian uses its VP across all categories, but others may use their own or delegated stake to participate to the reward scheme

Is it not the general approach that we can see every day? I guess I do not understand what you mean by that statement.

I am not sure I am following you now. Dmitri suggested to have a sep. pool dedicated to the translations and I said sure, you and others can do it, but Utopian won't, as it's VP is not meant to be dedicated solely to one category, but evenly spread across all of them.

All right, I think I get it now. Linking https://steemit.com/utopian-io/@dimitrisp/re-elear-re-dimitrisp-re-elear-talk-to-the-ceo-utopian-translations-category-20181114t215331118z because dimitrisp said there are some accounts used to support translators.

You got lost in the... translation 😂

Unfortunately, it seems that I did.

I’m working on my own response, which will be lengthier for sure, but since I was the one who lately raised the issue of scoring, especially with regards to post quality, I’ll quickly answer that part of your comment.

I would definitely score your latest contribution with a 78, assuming the translation is accurate of course... it’s all Greek to me! The post is definitely more engaging than average, I love the GIFs and I love the fact that it’s all your own words rather than blocks of quotations.
Which is, truly, the heart of the matter. You translated more than 1000 words on one of the more challenging projects currently available (which is still not above average difficulty, imo), did an excellent job and made a nice, personal, engaging presentation post. You couldn’t have done more. Which makes the fact that your team’s average scoring (as per the table posted under another comment on this thread) is higher than your 78 more than a little perplexing.
Which is not necessarily due to your LM “abusing” the system. It’s mostly the questionnaire’s fault, for being too generic and open to personal interpretation, which is some thing we figured out “in the field”.

That being said, I definitely agree that our biggest issue is the questionnaire itself. More on that later, though.

The average was over 78 because we were having difficult parts in more than 1 projects. When things stabilise again, you'll notice a drop on our average score (I guarantee that).

The difficulty issue is very troubling. If we are scoring The Curious Expedition as "Average", 90% of all the other projects are of High Difficulty. The Curious Expedition on average is "Very Easy" (an option we don't have on the questionnaire, not even "Easy"). And considering the language factor, in some languages it might be "Easy" instead of "Very Easy". But it's well established that the questionnaire is simply inefficient and when it changes, everything will be better.

Edit: Oh, and we don't have guidelines on what "Easy"/"Average"/"High"/"Higher" difficulties are (and no, the "school/medical text" markings on the questionnaire, are not helpful, see The Curious Expedition case I mentioned above :P )

I think communication can definitely be improved but I don't think that writing randomly on channels is a good idea because random text can quickly hide important things. It would make more sense to schedule fixed vocal chats and discuss by voice, we can think about a method to share documentation/points people want to discuss ahead of the meeting so to give everyone the time to think about what they want to say. I will reply to your other points later, first I want to give a chance to people to express their ideas and then I will join the discussion.

For the "random text" part, we could have a channel about small updates, with read only access for everyone except the CMs and the DaVinci staff.

I don't know if a vocal chat is feasible, considering timezone differences and our daily errands & work outside Utopian/Steemit. I know I can't commit to that as I recently quit my office job and right now I am a freelancer, so I have no standard working hours (to be honest, it's 22:40 here, and I should be working instead of writing walls of text 😂) but if it helps us transform our category and our community to something even better and I can put work aside (meaning it's not something urgent or a deadline is not approaching,) I will gladly do so.

Feel free to hit me up on discord if you want to discuss anything in private!

we definitely need to gather ideas in one place and improve the transfer of information, if you say something to another LM or CM but the DaVinci staff may not hear it..on the other hand we can't answer similar questions over and over again. So we can either work on shared documents where users can comment on or you elect some representatives for LM that will distill the major points and bring them to our attention.

Permission to upvote your comment with a bot? I don't have enough SP+VP for what it is actually worth.

if you like the idea we can definitely implement it. It's not too hard ;)

Sure, but let's wait and see what else will come out of thi "Call for Engagement" (as I'm sure there will be a lot of great suggestions) and we can do many things.

My offer to help in any way I can, is still on the table and when my app for the category is complete we can talk about what I can implement on it, to help you automate some stuff you might not have automated 😉

Thank you, we can definitely work something out

punishing translator X (who does a great job), just because translator Y (who didn't do as good of a job as X) didn't get an upvote

I don't really get what you mean by this. Could you elaborate?

Facts for the sentence, with random numbers:

  • Translator X got a score of 78 because they did a perfect translation.
  • Translator Y got a score of 40 because they didn't do great.
  • Utopian has to upvote a huge amount of contributions, so those with score <50 won't be upvoted today (and probably not even tomorrow)
  • 7 days passed, and Translator Y didn't get upvoted.

This happened when the new voting bot came into effect, some contributions didn't get upvoted, and even though the Utopian motto is "Upvotes are not guaranteed", people complained.

So there were many suggestions on how to change the questionnaire in a way so that everyone has a bigger chance for an upvote from Utopian.

While this would be awesome, it is: i) unsustainable, ii) impossible and iii) punishing those who go the extra mile to make an even better translation effort, because we have to minimise the maximum score a contribution can get if we want everyone to have a semi-guaranteed upvote. And it will become even more unsustainable if we start adding more languages.

I'm not saying I'm the genius that has it all figured out and I don't want to accuse anyone of being bad, as most people suggested things with the best intentions in mind. But we must adapt if we want this category to survive, understand that we may not get upvoted if our work is not perfect (and in some cases, even if it is perfect) and instead of complaining about other languages having better scores, discuss it with the respective LMs and try to raise your team's performance.

Damn it, I must stop writing walls of text. Sorry for this

So there were many suggestions on how to change the questionnaire in a way so that everyone has a bigger chance for an upvote from Utopian.

I guess this is what you meant by translator X getting punished? If so, then I agree with you and also think that the questionnaire should be used to separate the great quality contributions from the not so good ones instead of increasing the likelihood of everyone getting an upvote.

I guess this is what you meant by translator X getting punished?

Yes, but this is my interpretation of some of the things I saw (including reactions), so I might be wrong.

Edit: on a proposal I've seen, the maximum score can get to 86, and not to 100, for a perfect contribution

With the current steem price and considering that the translation category has an allotted 1.05% of the daily VP it means that only a couple of posts with a score of 80 could be voted for sure every day, everything else will depend on the spare VP from other categories. In times of high activity we had 17 contribution per day on average. Do you think people will continue translating if you only vote a few posts per day?

Wouldn't it be better if Utopian lowered the maximum upvote % a translation contribution can receive, instead of limiting the maximum score to 86 out of 100?

Same thing, in a better and more manageable way. What's easier to do: change an integer on the voting bot, or modify a whole questionnaire in the future if someone decides a translation should be able to get a bigger/smaller amount as a maximum?

Coin Marketplace

STEEM 0.18
TRX 0.15
JST 0.028
BTC 63177.41
ETH 2439.37
USDT 1.00
SBD 2.58