Announcement

Collapse
No announcement yet.

Firefox 117 Available With Local Automated Translation Support

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Luke
    replied
    Also if the government could find anyone regardless of security precautions, you would not see all those communiques for physical attacks on police stations/projects etc on anarchist websites posted over and over with nobody ever caught.

    Leave a comment:


  • Luke
    replied
    Google data BTW has sent a LOT of people to prison: the post Capitol storming geofence warrants are a well-known example.

    Leave a comment:


  • Luke
    replied
    Never assume the government can find you no matter what your security precautions: if THAT was true, all the posters of arson communiques over the Cop City project in Atlanta would have been caught. Instead, NONE have been caught and nobody has ever (as of 9-1-2023) been charged for one of the fires or for posting one of the communiques. Whatever you think of this project or of the resistance to it, what this proves is that serious computer counter-forensics can defeat a high priority, determined investigation with police and FBI involvement.

    Also never assume you have "nothing to hide." The example of Dutch records of religion used by the Nazis after they took over is one example. Another is that Dobbs abortion case in the US: before May 2022, nobody in the world of Internet services seriously expected their data to be used to lock up women for abortion, but that has not happened. Facebook not Google was the culprit this time around, but since this was a warrant, Google would have cooperated too. Defense is not to let Meta or Google know who you are or what you are doing.

    Google data BTW has sent a LOT of people to prison: many of the Jan 6 2021 US Capitol storming participants were found by geofence warrants to Google. The warrants sought all phone numbers etc that were logged into Google, had location sharing and location enabled, and were shown by GPS and/or "high precision" to be inside the US Capitol and not outside it. They then compared that to a list of all phone numbers expected to be inside the Capitol and got arrest warrants against the owners of all other phones on that list. Phone company tower data cannot tell being inside the Capitol from being on the Mall in front of the Capitol, possibly all the way to 3ed st NW. Those who did not have Google accounts, were not logged in, had location sharing off, or were on prepaid phones services with cash were never found except by other means.

    That was the US Capitol, next time could be gender-affirming care. Abortion cases are somewhat less likely w Google as they claim to destroy all such information coming in from phones, but if the GOP takes the White House in 2025 a "national security letter" could change that with zero notice.

    No matter WHO you are on the poliitical and economic spectrum, computer security and privacy directly affect your safety.

    Leave a comment:


  • oiaohm
    replied
    Originally posted by schmidtbag View Post
    By that logic, you shouldn't be visiting the site at all, because those packets are being sent over the Internet, and those packets can be intercepted. I figure GT is using SSL, so there's still some encryption. But yes, if all the details about your medical record are being logged by Google, that would be highly unacceptable.
    Yes intecption is risk but that why sites SSL between you and them. Yes google translate uses SSL but that does not solve the problems.
    (unofficial) Googletrans: Free and Unlimited Google translate API for Python. Translates totally free of charge. - ssut/py-googletrans


    Do note all the country locations. EU records that should stay in the EU chrome can end up exporting if you are using google translate built into chrome.

    Originally posted by schmidtbag View Post
    Hence my comment about PII. I don't think Google could be that dumb.
    Problem here
    France penalized the company for not properly disclosing to users how data is collected across its services to present personalized advertising.

    Google being that dumb before and got fined for it.
    With huge volumes of data to be processed with different countries laws its comes really easy doing online translation to goof up.

    The firefox route with a true local translation engine is the safer option all around.

    Remember you said highly not acceptable if google does log your medical record. There are lots of cases where google or any party running a browser integrated on line translation can end up being sent data they should not and then end up due to diagnostics of issues and the like end up long term storing records they should not have.

    Something to remember using full site translate inside chrome using google translate the URL has been stripped of this means google translate does not have the URL to go hang on this is coming from a location I should not be receiving data from due to country where the server is or due to what the data is.

    Leave a comment:


  • schmidtbag
    replied
    Originally posted by oiaohm View Post
    You are not getting it.

    ...

    Yes you are right it totally depends on what X entails. Lets say its your medical record with your home address and phone number and everything else. Remember the way the built in translate in chrome works happens to end up sending almost complete web pages you might not have the URL but you can still work out its a medical record and the like. There is a need with online translation for particular sites not to be translated.
    By that logic, you shouldn't be visiting the site at all, because those packets are being sent over the Internet, and those packets can be intercepted. I figure GT is using SSL, so there's still some encryption. But yes, if all the details about your medical record are being logged by Google, that would be highly unacceptable.
    EU laws the law does not kick in when you publicizing personal information the law kicks in when you are storing personal information you should not have. The way the EU Data protection laws are written are not compatible with online translation of particular data.
    Hence my comment about PII. I don't think Google could be that dumb.

    Leave a comment:


  • oiaohm
    replied
    Originally posted by schmidtbag View Post
    Agreed; I don't see a problem with local translate. But so long as Google isn't publicizing personal information, and, as long as they abide by the request to not track, I don't see a problem.
    You are not getting it.

    You can change your preferred language in Chrome. Chrome can also translate pages for you. Translate webpages in Chrome When you come across a page written in a language you don't understand

    This feature here. Is not 100 percent local translate as it sends copy off across internet and works on logged in pages.

    Originally posted by schmidtbag View Post
    Again: depends on what X entails. Google might not be at any legal risk if not enough data is being linked to identify someone.
    Yes you are right it totally depends on what X entails. Lets say its your medical record with your home address and phone number and everything else. Remember the way the built in translate in chrome works happens to end up sending almost complete web pages you might not have the URL but you can still work out its a medical record and the like. There is a need with online translation for particular sites not to be translated.


    Originally posted by schmidtbag View Post
    ​But so long as Google isn't publicizing personal information, and, as long as they abide by the request to not track, I don't see a problem.
    EU laws on personal information particular information storage without valid consent and reason is illegal. The reality is the way translate is implemented in chrome currently is going to end up with information google has no legal right to being sent to google translate for translation.

    EU laws the law does not kick in when you publicizing personal information the law kicks in when you are storing personal information you should not have. The way the EU Data protection laws are written are not compatible with online translation of particular data.

    Leave a comment:


  • schmidtbag
    replied
    Originally posted by oiaohm View Post
    This is the problem because what you said and highlighted is not correct for the google online translate built into chrome. Why I say lawsuit at some point is incoming on google at some point. EU does not move the fastest and will make sure they don't block user access to services this is why they fund the Firefox extension. The translate in google browser using online services happens to translate correctly secured sites that people has logged into.
    Unless Google is successfully using that sign-in through GT as an attempt to index their search engines, then they absolutely must be sued for that. Otherwise, I don't think they really can be sued over this.
    Local translate makes sense on a lot of sites. Google translate by website is different to google translate usage embedded in chrome browser. Yes Google translate by website can be blocked by robots.txt and requirement to login but Google online translate performed in chrome is not blocked by these things.
    Agreed; I don't see a problem with local translate. But so long as Google isn't publicizing personal information, and, as long as they abide by the request to not track, I don't see a problem.
    ​That the problem once a party is holding X information they cannot be sure never to leak X information and the more parties holding it the higher the leak risk becomes.
    Ultimately what matters is what X entails. If it directly ties back to you (outside of the internet) and is accessible to the average person, that is something I would be worried about. The possibility of that happening is >0%, but as stated before: I choose not to live my life in fear. Keep a low profile and no matter who gets the data, you're likely to be unaffected.
    With google account and other things with chrome the translated information being linked to you very simple to happen. Then think google chrome built in online translate support translating complete pages. Think different countries medical record sites have fixed formats. This is a problem once you start translating complete online pages from sites parties have logged its very simple as a online service hosting data you should not have in fact data that in many countries that it can be illegal for a company like google to have. Of course a person with a language miss match with the country they are in is going to want to use translate on government/medical sites with data companies like google legally should not have.
    Again: depends on what X entails. Google might not be at any legal risk if not enough data is being linked to identify someone. Think of it like an anonymous survey polling about very polarizing questions. Such surveys could collect data that, if not made anonymous, may either put the data collector or the survey takers (or both) in legal jeopardy. I could see Google not being successfully sued for possessing the diagnosis of someone's condition while otherwise not knowing who that person is, especially if that data never leaves Google's servers.

    Surely, Google is aware of PII.

    Leave a comment:


  • oiaohm
    replied
    Originally posted by schmidtbag View Post
    The likely reason why this hasn't occurred is because a properly secured site cannot be tracked. So if you have a login prompt, even if robots.txt says Allowed, none of these companies can collect any meaningful data. That's why whatever data is collected through GT isn't the same thing.
    This is the problem because what you said and highlighted is not correct for the google online translate built into chrome. Why I say lawsuit at some point is incoming on google at some point. EU does not move the fastest and will make sure they don't block user access to services this is why they fund the Firefox extension. The translate in google browser using online services happens to translate correctly secured sites that people has logged into.

    Local translate makes sense on a lot of sites. Google translate by website is different to google translate usage embedded in chrome browser. Yes Google translate by website can be blocked by robots.txt and requirement to login but Google online translate performed in chrome is not blocked by these things.

    Same problem happens with the google translate plugin with chrome and other vendors made online translate plugins for browsers.

    Originally posted by schmidtbag View Post
    So as long as GT doesn't make your medical records publicly accessible and doesn't tell anybody (including advertisers) what specifically what YOU have then the only information that is really collected is what site you've been to and perhaps the diagnosis, but that isn't necessarily linked to you specifically.
    ​That the problem once a party is holding X information they cannot be sure never to leak X information and the more parties holding it the higher the leak risk becomes.

    With google account and other things with chrome the translated information being linked to you very simple to happen. Then think google chrome built in online translate support translating complete pages. Think different countries medical record sites have fixed formats. This is a problem once you start translating complete online pages from sites parties have logged its very simple as a online service hosting data you should not have in fact data that in many countries that it can be illegal for a company like google to have. Of course a person with a language miss match with the country they are in is going to want to use translate on government/medical sites with data companies like google legally should not have.

    Local translate is a feature we need. Online translate for particular sites does need to be forbidden so that the company doing online translate does not end up storing data they have no legal right to have. Google at this stage with the google translate plugin and the feature in chrome to translate don't include means to say do not translate this site from the site side depends on user correctly opting out.


    Leave a comment:


  • schmidtbag
    replied
    Originally posted by treba View Post
    Well, history showed that political environments can change. You may find yourself in an environment where you suddenly *need* to hide something, potentially for very good reasons. At which point you might be happy if technical infrastructure is build in a way so it can not be easily abused.

    In the 1930s the Dutch kept records of religious affinity of their citizens. The Dutch government didn't abuse that info. But suddenly the Germans where in charge. Good for you if you believe that stuff like this can never happen again - but maybe imagine what some governments are ready to do to stop women to get an abortion, or what happens in states like China.
    I don't deny the possibility of such things happening, but I make a conscious choice to not live in fear, because that's not a good life. And for what it's worth: whatever dystopian world I end up in will be manageable. I pay may taxes, I have an honest middle-class job, I have a college degree, I have no health issues, I'm heterosexual, I live in a place of little political influence, and I have no religious affiliation. I'll be fine, but just because I'm in a relatively privileged position, that doesn't really change my point that most people, even those who don't check any of my boxes, don't have any immediate reason to concern over such companies and their data collection practices.

    Originally posted by oiaohm View Post
    Please note the EU requirement is because its bad for privacy. Lets say you are visiting the government medical site and you need translation of your own person medical history for some reason. Google translate in chrome end up ignoring robots.txt​,
    Is that what actually happens? Because that sounds so insecure that it would already be illegal without the EU adding a requirement.
    EU government sites are intentionally standalone for user privacy. So no google ads or facebook or... tracking bits. So how will google get tracking on those sites simple offer free translation in chrome using online services. Yes there is not a option to add to the sites robot.txt to say do not google online translate this. I can see the court case coming in future for Google over this in the EU.
    What kind of tracking are you referring to? Because to my knowledge, robots.txt is meant to prevent things like search engines, which publicize your site. It doesn't necessarily prevent anonymous data collection. So as long as GT doesn't make your medical records publicly accessible and doesn't tell anybody (including advertisers) what specifically what YOU have then the only information that is really collected is what site you've been to and perhaps the diagnosis, but that isn't necessarily linked to you specifically.
    Not that I'm defending this kind of data collection, but rather: I don't think it's as severe/incriminating as you're making it out to be.
    schmidtbag really what right does google and others have to track people on sites that have intentionally opts out of having these tracking bits. Translation is one of those things that when you think about it need to be able to run locally or done by the hosting server itself. Third party like google translation getting mixed up in this on many sites is not safe and lot cases should not be legal. Google translate should have entry to add to the sites robot.txt to say do not translate and other than robot.txt from the site nothing else will end up at google. Please note these EU governments sites that do not have tracking also normally have in the robot.txt do not crawl with a search engine.
    If there has been explicit opt-outs and it is tracking anyway, then as I have said: Google would already be ripe for a lawsuit. The likely reason why this hasn't occurred is because a properly secured site cannot be tracked. So if you have a login prompt, even if robots.txt says Allowed, none of these companies can collect any meaningful data. That's why whatever data is collected through GT isn't the same thing.
    Medical there is a big reason why you don't want advertising involved if possible. Think old snake oil sales men knowing what medical conditions person has and attempting to sell them products that don't work. Lot of medical stuff you want doctor prescribed for very good reasons.
    You almost had a very compelling point there, but here's the irony: in order for those snake oil salesmen to know your medical condition, that would have required you to either:
    A. Have already been diagnosed, in which case: a licensed doctor would have already given you options for treatment.
    B. You yourself have searched it and therefore, if you did your research appropriately, you would know what's snake oil and what isn't.
    Sure - there's bountiful gullible idiots out there who will try just about anything no matter how little scientific sense it makes. But, these people don't need salesmen or targeted ads: they're going to find pseudo-science crap regardless.
    The idea that you have nothing to hide is wrong. Advertiser getting particular information means they can then start targeting you with incorrect information that could trick you into getting something that is really bad for you. EU is funding this development because they see the safety problem. Particular things need to be private away from advertisers. Particular things don't need marketing mixed up in the mess.
    If I were a gullible and impulsive spender then sure. But, I have a hard enough time convincing myself to buy things I would really want or sometimes need. Again, you might argue that the gullible impulsive spenders need to be protected, but I know of such people and they had their issues long before the internet.

    Leave a comment:


  • oiaohm
    replied
    Originally posted by schmidtbag View Post
    Bad in the context of the EU, privacy, or both? Because I imagine using Google Translate to make a page legible is the least of people's privacy woes. GT is mediocre, and that matters because Google wants to collect your data because they want to know how to best target you for advertisements.
    Please note the EU requirement is because its bad for privacy. Lets say you are visiting the government medical site and you need translation of your own person medical history for some reason. Google translate in chrome end up ignoring robots.txt​,

    EU government sites are intentionally standalone for user privacy. So no google ads or facebook or... tracking bits. So how will google get tracking on those sites simple offer free translation in chrome using online services. Yes there is not a option to add to the sites robot.txt to say do not google online translate this. I can see the court case coming in future for Google over this in the EU.

    schmidtbag really what right does google and others have to track people on sites that have intentionally opts out of having these tracking bits. Translation is one of those things that when you think about it need to be able to run locally or done by the hosting server itself. Third party like google translation getting mixed up in this on many sites is not safe and lot cases should not be legal. Google translate should have entry to add to the sites robot.txt to say do not translate and other than robot.txt from the site nothing else will end up at google. Please note these EU governments sites that do not have tracking also normally have in the robot.txt do not crawl with a search engine.

    Medical there is a big reason why you don't want advertising involved if possible. Think old snake oil sales men knowing what medical conditions person has and attempting to sell them products that don't work. Lot of medical stuff you want doctor prescribed for very good reasons.

    Medical in many countries its the case but its not the only case.

    The idea that you have nothing to hide is wrong. Advertiser getting particular information means they can then start targeting you with incorrect information that could trick you into getting something that is really bad for you. EU is funding this development because they see the safety problem. Particular things need to be private away from advertisers. Particular things don't need marketing mixed up in the mess.

    EU has a lot of strict data rules and lots of those rules should be global because the reasoning behind the rules are based on case examples of how the existing system before the rules end up doing bad things.

    Leave a comment:

Working...
X