Crowdin for translations

[asmeurer]

We should look at https://crowdin.com/ for translations. NumPy is going to use it for their new website. I think they have a free tier for open source projects.

[certik]

Great idea. (The first flag they show is Czech, how did they know?)

[goanpeca]

We also use Crowdin for Spyder and JupyterLab

I would say go for it :-) !

[asmeurer]

I think we just need someone to set it up. We already have the translation infrastructure set up for this repo, so it shouldn't be hard. I think it should be fine to have it automatically commit translations to the repo.

[goanpeca]

I think we just need someone to set it up.

Do you have a bot account for this so that PRs are not committed by whoever sets the integration (but the bot) ?

[asmeurer]

Yes, we can use @sympy-bot. Although that raises the question, how are the translators attributed in the repo?

[goanpeca]

Although that raises the question, how are the translators attributed in the repo?

The list is on crowdin, maybe we should be copying that information in the repo.... hmmm with a bot :-p

[goanpeca]

@pearu?

[comradekingu]

@asmeurer If you read the terms and conditions you will notice it is not. The plan requires sub-licensing strings into the closed translation memory of Crowdin. For other damning details, see jitsi/jitsi-meet#5056 (comment)
Use Weblate instead, which actually is and has a gratis plan for libre software projects :)

[asmeurer]

@comradekingu can you refer to which part of the terms and conditions you are talking about? I don't see it as an issue if crowdin gets a license to use the translations done for the SymPy website, since they would already be liberally licensed. We don't actually have a license file on this repo, but I would say it should be licensed BSD, the same as SymPy itself. So as long as their terms allow us to do that, I don't see a problem.

I don't have a strong opinion myself as to which service we use, and if someone wants to take the time to set one, that's the one we should use. I only speak one language myself, so I wouldn't participate in the translations. I've heard from people that they like the crowdin interface, so that is why I suggested it. If weblate or some other service seems better, we can set that up instead.

[comradekingu]

@asmeurer
It is impossible to get any overview with the Crowdin interface, and the context to what string is what, and where it enters the fray is lost. That is the interface you get, with no alternative ones to pick from. This is why translations on Crowdin are not consistent, bar maybe very short ones. It is also much harder to get to the source strings from there. Etc. etc.

It is underhanded to pretend sublicensing "supports" libre software. As the case may be, there isn't any license for this repo, but in the case of "numphy" the distinction (depending what you think about the text) is with this part of the 3-clause BSD license: (which would otherwise need to be upheld)

" * Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer."

think what you will about just taking BSD code that way without contributing back.

In any event the term "Crowdin for Open Source" "We support the open source community" libre software package turns out to be only for "non-profit" projects, meaning no donations, no nothing. This even undermines libre software as a concept, by creating confusion as to what libre software is. In the very least it isn't how to grow one. Crowdin itself neither is libre software, nor keeps a translation memory that is. Instead it employs every shady practice under the sun to sell and mine data. Not OK.

> **web beacons on all users**.

Here are some direct clippings of all other relevant parts of
https://support.crowdin.com/privacy-policy/.
(My comments only to be found betwixt those, everything else verbatim, in the order it appears, minus irrelevant sections. I have made it easier by cutting it down in bold, which doesn't change the content, presented as is).

---

We may disclose information to third parties if you consent to us doing so, as well as in the following circumstances:

(OK, what constitutes consent, and what does that entail. Can the service be used at all without it?)

We work with third party service providers who provide website, application development, hosting, maintenance, and other services for us. These third parties may have access to, or process Personal Data or Client Data as part of providing those services for us.
We may make certain automatically-collected, aggregated, or otherwise non-personally-identifiable information available to third parties for various purposes, including (i) compliance with various reporting obligations; (ii) for business or marketing purposes; or (iii) to assist such parties in understanding our Clients’, Users’ and Visitors’ interests, habits, and usage patterns for certain programs, content, services, and/or functionality available through the Service

We may disclose Personal Data or other information if required to do so by law or in the good-faith belief that such action is necessary to comply with applicable laws, in response to a facially valid court order, judicial or other government subpoena or warrant, or to otherwise cooperate with law enforcement or other governmental agencies.

We also reserve the right to disclose Personal Data or other information that we believe, in good faith, is appropriate or necessary to (i) take precautions against liability, (ii) protect ourselves or others from fraudulent, abusive, or unlawful uses or activity, (iii) investigate and defend ourselves against any third-party claims or allegations, (iv) protect the security or integrity of the Service and any facilities or equipment used to make the Service available, or (v) protect our property or other legal rights, enforce our contracts, or protect the rights, property, or safety of others.

(Wait for it)

Information about Users and Visitors, including Personal Data, may be disclosed and otherwise transferred to an acquirer, successor or assignee as part of any merger, acquisition, debt financing, sale of assets, or similar transaction, as well as in the event of an insolvency, bankruptcy, or receivership in which information is transferred to one or more third parties as one of our business assets and only if the recipient of the User or Visitor Data commits to a Privacy Policy that has terms substantially consistent with this Privacy Policy.

Client Data may be physically or electronically transferred to an acquirer, or successor or assignee as part of any merger, acquisition, debt financing, sale of assets, or similar transaction, as well as in the event of an insolvency, bankruptcy, or receivership in which information is transferred to one or more third parties as one of our business assets, for the sole purpose of continuing the operation of the Service, and only if the recipient of the Client Data commits to a Privacy Policy that has terms substantially consistent with this Privacy Policy

You may decline to share certain Personal Data with us, in which case we may not be able to provide to you some of the features and functionality of the Service.

The Service may contain features or links to web sites and services provided by third parties. Any information you provide on third-party sites or services is provided directly to the operators of such services and is subject to those operators’ policies, if any, governing privacy and security, even if accessed through the Service. We are not responsible for the content or privacy and security practices and policies of third-party sites or services to which links or access are provided through the Service.

Interest based advertising is the collection of data from different sources and across different platforms in order to predict an individual’s preferences or interest and to deliver to that individual, or his/her computer, smart phone or tablet, advertising based on his/her assumed preference or interest inferred from the collection of data pertaining to that individual or others who may have a similar profile or similar interests.

We work with a variety of third parties to attempt to understand the profiles of the individuals who are most likely to be interested in the Crowdin products or services so that we can send them promotional emails, or serve our advertisements to them on the websites and mobile apps of other entities.

These third parties include: (i) advertising networks, which collect information about a person’s interests when that person views or interacts with one of their advertisements; (ii) attribution partners, which measure the effectiveness of certain advertisements; and (iii) business partners, which collect information when a person views or interacts with one of their advertisements.

In collaboration with these third parties, we collect information about our customers, prospects and other individuals over time and across different platforms when they use these platforms or interact with them. Individuals may submit information directly on our Sites or on platforms run by third parties, or by interacting with us, our advertisements, or emails they receive from us or from third parties. We may use special tools that are commonly used for this purpose, such as cookies, pixels, and similar technologies. We may have access to databases of information collected by our business partners.

The information we or third party collect enables us to learn what purchases the person made, what ads or content the person sees, on which ads or links the person clicks, and other actions that the person takes on our Sites, or in response to our emails, or when visiting or using third parties’ platforms.

We, or the third parties with which we work, use the information collected as described above to understand the various activities and behaviors of our customers, Site visitors and others. We, or these third parties, do this for many reasons, including: to recognize new or past visitors to our Sites; to present more personalized content; to provide more useful and relevant ads - for example, if we know what ads you are shown we can try not to show you the same ones repeatedly; to identify visitors across devices, sales channels, third party websites and Sites, or to display or send personalized or targeted ads and other custom content that is more focused on a person’s perceived interest in products or services similar to those that we offer.

Our interest-based ads may be served to you in emails or on third-party platforms. We may serve these ads about our products or services or send commercial communications directly ourselves or through these third parties. Crowdin does not provide any personal information to the third party sites that display interest-based ads promoting Crowdin.

(***Just have to interject here to differentiate between personal information and personally identifiable information, read on)

We allow third parties with which we have a separate agreement to use cookies and other technologies to collect information about your use of the Site. These third parties include (i) business partners, which collect information when you view or interact with one of their advertisements on the Site; and (ii) advertising networks, which collect information about your interests when you view or interact with one of their advertisements.

The information gathered by these third parties is used to make predictions about your interests or preferences so that they can display advertisements or promotional material on this Site and on other sites across the Internet tailored to your apparent interests.

The business partners and advertising networks that serve interest-based advertisements on the Services have limited access to a small amount of information about your profile and your device, which is necessary to serve you advertisements that are tailored to your apparent interests. It is possible that they may reuse this small amount of information on other sites or services.

(Bingo***)

We do not share with these third parties any information that would readily identify you (such as email address); however, these third parties may have access to information about your device (such as IP or MAC address). We do not have access to, or control over, the technologies that these third parties may use to collect information about your interests, and the information practices of these third parties are not covered by this Privacy Notice. Other than as discussed in this document, we have no control over these third parties.

(Interjecting again, e-mail address is identifiable, but IP and MAC isn't? Be the judge of this information.)

functionality cookies - These cookies allow the website to remember choices you make (such as your user name, language or the region you are in) and provide enhanced, more personal features. For instance, a website may be able to provide you with local weather reports or traffic news by storing in a cookie the region
in which you are currently located. These cookies can also be used to remember changes you have made to text size, fonts and other parts of web pages that you can customise. They may also be used to provide services you have asked for such as watching a video or commenting on a blog. The information these cookies collect may be anonymised and they cannot track your browsing activity on other websites.

(I'll let the reader be the judge of whether this in fact this grants a unique fingerprint that can track you across websites)

behaviourally targeted advertising cookies - These cookies are used to deliver adverts more relevant to you and your interests. They are also used to limit the number of times you see an advertisement as well as help measure the effectiveness of the advertising campaigns. They are usually placed by advertising networks with the website operator’s permission. They remember that you have visited a website and this information is shared with other organisations such as advertisers. Quite often targeting or advertising cookies will be linked to site functionality provided by the other organisation.

Protecting the privacy of young children is especially important. Our Service is not directed to children under the age of 18, and we do not knowingly collect Personal Data from children under the age of 18 without obtaining parental consent. If you are under 18 years of age, then please do not use or access the Service at any time or in any manner. If we learn that Personal Data has been collected on the Service from persons under 18 years of age and without verifiable parental consent, then we will take the appropriate steps to delete this information.

(Remember the part about good-faith earlier?)

The Service is not intended to be used by minors, and is not intended to be used to post content to share publicly or with friends. To the extent that a minor has posted such content on the Service, the minor has the right to have this content deleted or removed using the deletion or removal options detailed in this Privacy Policy. If you have any question regarding this topic, please contact us as indicated in the “How to Contact Us” section. Please be aware that, although we offer this deletion capability**,** the removal of content may not ensure complete or comprehensive removal of that content or information.

Although we may allow you to adjust your privacy settings to limit access to certain Personal Data, please be aware that no security measures are perfect or impenetrable. We are not responsible for circumvention of any privacy settings or security measures on the Service. Additionally, we cannot control the actions of other users with whom you may choose to share your information. Further, even after information posted on the Service is removed, caching and archiving services may have saved that information, and other users or third parties may have copied or stored the information available on the Service.

We may transfer, process and store Personal Data we collect through the Services in centralized databases and with service providers located in the U.S. The U.S. may not have the same data protection framework as the country from which you may be using the Services.

(Take heed of why this needs to be pointed out. Do any particular US laws come to mind? Bingo.)

The Service is hosted in the United States or Germany. Regardless of the database being hosted in the EU, if you choose to use the Service from the European Union or other regions of the world with laws governing data collection and use that may differ from U.S. law, then please note that you may be transferring your Client Data and Personal Data outside of those regions to the United States for storage and processing by our service providers listed in the Crowdin Terms of Service.

(But nobody is on that list, right?)

Crowdin is largely unaware of what Client Data is actually being stored or made available by a Client or User to the Service and does not directly access such Client Data except as authorized by the Client, or as necessary to provide Services to the Client and its Users.

(It is almost as if the EU GDPR has provisions for this)

Because Crowdin does not collect or determine the use of any Personal Data contained in the Client Data and because it does not determine the purposes for which such Personal Data is collected, the means of collecting such Personal Data, or the uses of such Personal Data, Crowdin is not acting in the capacity of data controller in terms of the European Union’s General Data Protection Regulation (Regulation (EU) 2016/679, hereinafter “GDPR”) and does not have the associated responsibilities under the GDPR. Crowdin should be considered only as a processor on behalf of its Clients and Users as to any Client Data containing Personal Data that is subject to the requirements of the GDPR. Except as provided in this Privacy Policy, Crowdin does not independently cause Client Data containing Personal Data stored in connection with the Services to be transferred or otherwise made available to third parties, except to third party subcontractors who may process such data on behalf of Crowdin in connection with Crowdin’s provision of Services to Clients. Such actions are performed or authorized only by the applicable Client or User.

(And it is almost like Crowdin has noticed)

The Client or the User is the data controller under the Regulation for any Client Data containing Personal Data, meaning that such party controls the manner such Personal Data is collected and used as well as the determination of the purposes and means of the processing of such Personal Data.

Crowdin is not responsible for the content of the Personal Data contained in the Client Data or other information stored on its servers (or its subcontractors’ servers) at the discretion of the Client or User nor is Crowdin responsible for the manner in which the Client or User collects, handles disclosure, distributes or otherwise processes such information.

Please revisit this page periodically to stay aware of any changes to this Policy, which we may update from time to time. If we modify the Policy, we will make it available through the Service, and indicate the date of the latest revision, and will comply with applicable law. Your continued use of the Service after the revised Policy has become effective indicates that you have read, understood and agreed to the current version of the Policy.

(What a complete spaghetti cluster of vagueness, and outright scandalous terms and practices. Notice the pattern of always mixing legitimate concerns with those that are not, the pattern of how they are referenced, and how connected parts are hidden in different categories, under headers that seem like they concern other concerns entirely. Look how the content of sentences is padded to seem less alarming, and how fill like "may", etc. just happen to appear exactly everywhere the alarming action is carried out.)

There is also the https://support.crowdin.com/cookies/ and https://downloads.crowdin.com/docs/DPA-singed.pdf

TL;DR You can't use the service without accepting, that your info is sold according to ability, for the purpose personal tracking, etc., you leak info to the NSA en-masse, all of the above happens live, and it doesn't matter if info is removed afterwards, and if it did, it is not promised that it will be gone, and the GDPR supposedly can't help you, because Crowdin doesn't know what goes on on their platform (they say), and doesn't think so (good luck with that), and you are supposed to be knowledgeable about both all of the above, and continually know what is in this document, which can change at any time, based on a date timestamp at the end.)

[asmeurer]

I disagree that crowdin is sublicensing the translations. A person contributing a translation is licensing it under the license for this repo, as well as under the crowdin terms that they agree to when signing up.

I don't particularly care if Crowdin uses the translations for its own ML or whatever. Anyone could take the translations from the website itself or from the sources in this repo and do that anyway. It makes sense that they would explicitly require permissions via their ToS to do this so they don't have to worry about the notices part of the BSD license. As long as Crowdin allows the translations made on its site for the SymPy website to be contributed to the SymPy website under the respective open source license, I don't see any issues.

In any event the term "Crowdin for Open Source" "We support the open source community" libre software package turns out to be only for "non-profit" projects, meaning no donations, no nothing.

The word "non-profit" doesn't appear anywhere in the crowdin terms of service. On this page, it only says "you do not have any commercial products related to the open-source project you are requesting a license for," which only means that the translations for the free "open source" tier would only be used for an open source project (which would be the case for the SymPy website).

TL;DR You can't use the service without accepting, that your info is sold according to ability, for the purpose personal tracking, etc., you leak info to the NSA en-masse, all of the above happens live, and it doesn't matter if info is removed afterwards, and if it did, it is not promised that it will be gone, and the GDPR supposedly can't help you, because Crowdin doesn't know what goes on on their platform (they say), and doesn't think so (good luck with that), and you are supposed to be knowledgeable about both all of the above, and continually know what is in this document, which can change at any time, based on a date timestamp at the end.)

What info would crowdin be collecting that isn't already public?

[comradekingu]

@asmeurer What alternative is there to using Crowdin if you want to contribute then? It snatches everything that enters the codebase on conditions of incredulity, subject to a court far outside the realms of the common translator.

I don't think translators really know what they are contributing to in that regard, and I can at-least attest to my motivation being that of only helping libre software projects. Then I actually read the terms and conditions/EULA. No you can't just lift the current website, as its basis constitutes a work, which is subject to copyright. Being a copyright holder depending on license chosen, (of which there is none now) means contributing.

However issueless you may deem the practice of shifting the axis of for profit motives, I think we can agree it is not optimal in its outcome. It does bring into question what alternatives could and do exist.

It says "If you're building awesome non-profit projects that could use the power of Crowdin, we're happy to help." and even interpreted in the meaning of related products, it means no commercialization. No T-shirts, no paid support, no support tiers. Transifex is already big enough to hold projects to the candle with those type of terms, and they do. Question why that (subject to change) text is even in the document at this point. As opposed to nothing, it serves a purpose. Look at the alternative endeavour, and you find no such clause. I see no issue in avoiding that pitfall. Using Crowdin means you can't ever fork it. If there is no issue moving platforms, sooner rather than later seems better.

If you click the "details" blurb above you will see Crowdin collects just about everything, and employs some spurious definitions and language to support it, in spaghetti fashion. Or you can load the site and plainly see.
From what I remember there is a list somewhere, and of course every single interaction is subject to respective terms and conditions. In the aggregate you have a valuable dataset that is hard enough to get even the basic of insight into. Again, the alternative is there and a much less complex proposal to similarly not take issue with in the current and foreseeable future.