Stats against Weblate spam #1
Comments
|
Repeatedly subjective spammed without data or citation. Transifex and Crowdin are bad. Weblate is the best. SponsorBlock has launched DeArrow. The goal is to correct hype. Thus, People are sensitive to this.
Clearly rejected. Changing translation platforms could result in loss of existing history and contributors.
Then introduce Weblate. But the fact is that Weblate separates listings and multiple languages. The original service (Crowdin) combined these into one window.
Weblate has no bugs? I have also filed many bug reports for Weblate.
It is countered that the ownership issues he mentions also occur in Weblate.
Rejected. Crowdin's history listing and voting system is easy to use. Details. Strange translations are eliminated with a negative rating in Japanese Translation of SponsorBlock. Crowdin seems to have a lot of users. I don't know what he is based on any data.
|
|
Hello @maboroshin, Benjamin from Weblate here. I would like to offer essential information. The Weblate team do not have anything to do with this effort. I got a notification about your comment in blueman-project/blueman#1149 (comment) as I contributed to the discussion there. It guided me here, and I left a new comment there, too. Suggesting a tool in an open-source project repository does not seem like a spam. The community and the maintainers can reject the suggestion, close the issue, even lock it or mark it as SPAM if they feel it. Keeping a good attitude in discussions can be a tricky discipline sometime. Everyone should be respectful, no one should be forced to do or use something; it is not about winning. If someone’s offensive, you can tell them and/or report them. @comradekingu is a translator, Weblate user, a fan of open source. They do not work for Weblate, and we never asked him to “advertise” Weblate this way; we do not advertise Weblate at all. Everyone is free to use Weblate as it is a libre software. We welcome new open source projects the same way we welcome new commercial customers – by a word of a current satisfied Weblate user. Signing commits by e-mail addresses is a feature, it is how Git works. Open repositories are a feature of open-source software. To top of that:
If you want, @maboroshin, we can connect and chat about anything Weblate, this topic included. Our care (at) weblate dot org address seems like a good place to start for that. We also use Discussions. |
|
from the initial post:
I am not going to fact-check all the issues in this comparison, but this is not correct. Access control is available in a free plan, but yes, you can’t limit the visibility of the project; it’s open-source after all. Any project owner/admin can prevent a user from contributing to their project, etc. More here: https://docs.weblate.org/en/latest/admin/access.html#project-access-control This seems more like a collection of things randomly scraped from the internet than a reliable source. The information is incorrect, and we are not affiliated with it. Weblate numbers in the sheet are not correct, you can’t count all projects using Weblate. |
SpamI understand it. I guess it is not the Weblate team that is doing this. If there is a monetary award, it is prohibited by the state as stealth marketing. LeakAt the end of 2022 we can choose not to have Weblate publish our e-mails. This is due to the following concerns about leaks. However, published emails are still published on each Github commit. Under GDPR, email addresses are personal data. Anyone can view that personal data. Personal data is exposed outside of Weblate and can't be erased. pappasadrian said
p0358 said
(PS)
There is no mention about disclosure of email addresses. Before 2022, no one has selected the setting "make my email address public". There would have been no explicit consent. (End of PS) Access controllPlease read carefully the page you referred to.
Reliable sourceI'm sorry. Please let me know if you have any reliable sources. I will replace it. |
|
Feels to me like you are trying to solve problems when there are no problems. Thus, I am unable to help.
|
|
Leak: Anyone can see contributor's email address. Some call it a feature, some call it a leak. I quoted others' statements above. Access control: The documentation should say that access control is available even with the free Libre plan. With the current documentation, many administrators would also conclude that access control is not possible. Reliable source: It doesn't seem to have the information I need on it. I will try to search again later . PS It was possible to block users from the project after mid-2021 (version 4.7). It was mentioned in another heading scrolling down 3 pages. I will correction to the description. I appreciate your explanation. |
|
No, only you call it a leak when it isn't. |
|
I'm not the only one. already cited. Access control is a feature found in Transifex, Crowdin, and Weblate. Is this a feature that is not in demand? It's a feature used: WeblateOrg/weblate#3548 WeblateOrg/weblate#8470 If you know the correct information published, I will change to it.
All data has been changed to Linkedin data. Compared to rocketreach, the number of employees in Weblate remains unchanged at 4. The overall description is the same. |
|
Why would it be strange for the last few libre software projects actually using C or TX may not be interesting in putting in effort to move? Crowdin for example has lots of employees that only do marketing, while pretending otherwise. To deal with what seems the remaining question you have: |
|
Since you keep linking to this as "objective" without detailing it is your own creating, and calling it "research". The link is right where you posted it, but it bears repeating
NightscoutFoundation/xDrip#3420 (comment) You evidently don't know how the platforms work, and that is fine. You haven't read the terms and conditions, cookie policy, etc. of either TX or C, and instead put a question-mark for their respective entries in "Leakage of personal data". https://support.crowdin.com/privacy-policy/ It is a spaghetti mess of multiple documents and potentially indemnifying statements mixed in with a lot of caveats to explain the actual behavior in what "may" happen, by design. It doesn't help that something only available as a service makes one completely dependent on these terms, and their continual changes. for contrast, for those playing at home https://weblate.org/en/terms/ |
|
It is good news that the number of strings has increased to more than 10,000. But certainly in 2023 it's 10,000 strings for Libre softwares. If there is a fact of personal data leakage, please tell me the data. I can see the email addresses of Weblate users on GitHub. Issues were continually being made regarding Weblate's handling of personal information. Because, Weblate was not strictly compliant with GDPR. Recent one: Even the cookie policy will be clearly stated in order to comply with GDPR.
There is no transparency as to what Weblate is actually doing. Transifex and Crowdin are strict with GDPR. But, You fear that the document is long according to GDPR. You are fantasy and not specific. |
You usually end up with that kind of mess as the legal environment evolves, and you try to update the terms without rewriting the whole of them. This doesn't necessarily have to be a bad thing. On the other side, we at Weblate have ignored some legal changes in the past years, and we will soon have a wholly new terms of service because of that. These will be much more complex than the current ones, even if I fight with the lawyers to make them as simple as possible.
We've moved from counting source strings to hosted strings, so these numbers have changed, but the numbers also count different things. We allowed 10,000 apples before, and now we allow 160,000 cherries. It might be better for some projects, but it also might be worse for others, but we believe it's a more fair way of estimating translation project size. On the other side, we would rather not use the words metric (what Transifex and Crowdin use) because it's pretty ambiguous for many languages.
You can only choose not to use the non-essential cookies. Weblate uses essential cookie only (for session/authentication). These services have a complex cookie policy because they heavily depend on third-party cookies. There is no need for a complex cookie policy when the service uses a single cookie (the second cookie we've used so far is just about to be removed in WeblateOrg/weblate#12383). |
Spams are not related to the Weblate team. A reference to the data is needed. Because fantastic subjective expressions are used extensively. We need a way to quantify this.
Nominal
(2018)
Nominal
Nominal
Nominal
Actual
Actual
Nominal
Nominal
67,413
Nominal
+Contributors
words
strings in 2024
was 10,000 in 2023Details
Service down
Many times in a month Weblate will not allow you to edit anything. maintenance mode for several hours for the whole Weblate.
Blocks of individual projects that can last as long as several days. It will continue until the administrator takes action. In some cases, it has lasted more than a month. Please search ```The translation is temporarily closed for contributions due to maintenance, please come back later'''
With other services I don't encounter this. In other services, at least, the frequency will be much less frequent.
Note for numbers
Description
Wordpress and Launchpad have translation guidelines. A large number of translations can be done with uniformity. Depending on the project, Transifex and Crowdin require approval to participate. And, These two are open to participants who are at least 18 years old according to the terms and uses. Many Weblate projects are open to everyone including kids.
Leakage of personal data
GDPR defines email addresses as personal data. Weblate has already output mail addresses to Github and it is difficult to erase all of it. This is worse when combined with the GDPR right to erase data. guardianproject/orbot#858 See also detail
The project has been offered to migrate from Transifex to Weblate and the contributors' email addresses have been leaked. blueman-project/blueman#1149
Revision
The text was updated successfully, but these errors were encountered: