4 ms·

Mozilla, you are 100% completely and absolutely out of touch. The problem is not that a study was conducted incorrectly! The problem is that software was secret

Mozilla, you are 100% completely and absolutely out of touch. The problem is not that a study was conducted incorrectly! The problem is that software was secretly bundled with Firefox, software that is at the very least bloat and might be considered adware -are you paid for this? Was it a favor? Part of a non financial partnership/promotion? You don't see the problem here, no, you plan on doing this again. With the only change being that you will try to answer a specific question. You could not have mitigated the uproar by "trying to answer a specific question" and "appropriately naming the 'study'". The biggest issue is that you are unable to realize that.

maaaats 9y ago
This post wasn't about answering those questions, that they have done already elsewhere (some of it linked in the article, even). This was about how they will make sure not to use SHIELD to do the same in the future.
- Svip 9y ago
  Hell, that post[0] is linked in the first paragraph. [0] https://blog.mozilla.org/firefox/update-looking-glass-add/
- pbhjpbhj 9y ago
  3 times in, what, a year they've now tried deceptively foisting bundled changes (with commercial links) ... whoever is in charge clearly didn't get it and just thinks it's about implementation details.
  - maaaats 9y ago
    What 3 times are we talking about?
    - Fnoord 9y ago
      Pocket integration was in 2015 (acquisition was in Feb 2017 [1] as linked in this article as well) so not last year, but the Cliqz debacle was last year. Assuming Looking Glass is the most recent one I'm not sure about #3. Describing Pocket as being #3 within a year is dishonest IMO. [1] https://blog.mozilla.org/blog/2017/02/27/mozilla-acquires-pocket/ https://blog.mozilla.org/blog/2017/02/27/mozilla-acquires-po...
      - pbhjpbhj 9y ago
        Ok, so if it's in 2 years that makes it fine? Pocket, Hello/Telefonica, Pocket again (or was it Hello, my recollection is hazy, they forced the icons back for users who's removed them), Mr Robot, ... I apologise if it causes some harm that I have imperfect recollection.
        Fnoord 9y ago
        > Ok, so if it's in 2 years that makes it fine? Not what I was arguing. The purpose of my post was helping out recollecting the mistakes Mozilla made. I don't remember the Hello issue. You're also forgetting the Cliqz debacle. Sure, it was German-specific, but still. I don't think it is fine, and it is good Mozilla is being criticised. However, the alternatives are arguably much worse.
- zelos 9y ago
  Come on, this doesn't trigger your weasel words detector? >An ‘experiment’ that does not capture any data is not an experiment at all. >In retrospect, not capturing data was a strong indicator that this was not a good SHIELD study candidate, so we’re making sure we’re going to specifically evaluate future studies So a commercial add on that _does_ collect data is ok?
pdpi 9y ago
> You could not have mitigated the uproar by "trying to answer a specific question" and "appropriately naming the 'study'". The biggest issue is that you are unable to realize that. That’s not how I read it. What they’re saying here is “the fact that it didn’t try to answer a question is a red flag unto itself”. If it had tried to answer a question, it would’ve collected data, and the usual privacy process would’ve kicked in hard: is this an appropriate question to ask, are the changes narrowly tailored to ask that question, are we handling the data appropriately? On those counts, a version of Looking Glass that collects data would never have rolled out. Instead, by not collecting data, they found themselves answering an overly narrow question: does this impact user privacy? No, it does not! Naming really was adding more fuel to the fire though. It’s a fine line between a surprise/Easter egg and being outright deceitful, and the cock up with the SHIELD test burnt through all the goodwill they might have here.
- nlawalker 9y ago
  >> Instead, by not collecting data, they found themselves answering an overly narrow question: does this impact user privacy? What makes them out of touch is that these were apparently the only criteria that needed to be met. As opposed to, say, "how would people feel if we took our software, which occupies a space in their workflow that demands a lot of trust, and auto-updated it with a marketing campaign for a TV show?" I mean, if they wanted to answer the question "how many of our users like Big Bird?", and they crafted a pop-up survey to be delivered through this mechanism that did not impact user privacy, would that get the green light? It's not just about respecting user privacy and safeguarding data, it's about adhering to expectations. The fact that there was apparently not a single human in the loop with the power to stop this from rolling out that recognized this is concerning, and the fact that there's no mention of it in this blog post is doubly so. What people want to hear is not just "we will safeguard your data", it's "we will never waste your time or introduce any kind of risk by pushing things like marketing campaigns into your trusted software."
  - pdpi 9y ago
    Yes, absolutely. Hence my describing it as an overly narrow question. My point is precisely that, in going the SHIELD route, they selected for people who would ask that narrow question, because that's what they're usually concerned with, rather than the bigger questions they should've been asking.

maaaats 9y ago

This post wasn't about answering those questions, that they have done already elsewhere (some of it linked in the article, even). This was about how they will make sure not to use SHIELD to do the same in the future.

Svip 9y ago
Hell, that post[0] is linked in the first paragraph. [0] https://blog.mozilla.org/firefox/update-looking-glass-add/
pbhjpbhj 9y ago
3 times in, what, a year they've now tried deceptively foisting bundled changes (with commercial links) ... whoever is in charge clearly didn't get it and just thinks it's about implementation details.
- maaaats 9y ago
  What 3 times are we talking about?
  - Fnoord 9y ago
    Pocket integration was in 2015 (acquisition was in Feb 2017 [1] as linked in this article as well) so not last year, but the Cliqz debacle was last year. Assuming Looking Glass is the most recent one I'm not sure about #3. Describing Pocket as being #3 within a year is dishonest IMO. [1] https://blog.mozilla.org/blog/2017/02/27/mozilla-acquires-pocket/ https://blog.mozilla.org/blog/2017/02/27/mozilla-acquires-po...
    - pbhjpbhj 9y ago
      Ok, so if it's in 2 years that makes it fine? Pocket, Hello/Telefonica, Pocket again (or was it Hello, my recollection is hazy, they forced the icons back for users who's removed them), Mr Robot, ... I apologise if it causes some harm that I have imperfect recollection.
      - Fnoord 9y ago
        > Ok, so if it's in 2 years that makes it fine? Not what I was arguing. The purpose of my post was helping out recollecting the mistakes Mozilla made. I don't remember the Hello issue. You're also forgetting the Cliqz debacle. Sure, it was German-specific, but still. I don't think it is fine, and it is good Mozilla is being criticised. However, the alternatives are arguably much worse.
zelos 9y ago
Come on, this doesn't trigger your weasel words detector? >An ‘experiment’ that does not capture any data is not an experiment at all. >In retrospect, not capturing data was a strong indicator that this was not a good SHIELD study candidate, so we’re making sure we’re going to specifically evaluate future studies So a commercial add on that _does_ collect data is ok?

pdpi 9y ago

> You could not have mitigated the uproar by "trying to answer a specific question" and "appropriately naming the 'study'". The biggest issue is that you are unable to realize that. That’s not how I read it. What they’re saying here is “the fact that it didn’t try to answer a question is a red flag unto itself”. If it had tried to answer a question, it would’ve collected data, and the usual privacy process would’ve kicked in hard: is this an appropriate question to ask, are the changes narrowly tailored to ask that question, are we handling the data appropriately? On those counts, a version of Looking Glass that collects data would never have rolled out. Instead, by not collecting data, they found themselves answering an overly narrow question: does this impact user privacy? No, it does not! Naming really was adding more fuel to the fire though. It’s a fine line between a surprise/Easter egg and being outright deceitful, and the cock up with the SHIELD test burnt through all the goodwill they might have here.

nlawalker 9y ago
>> Instead, by not collecting data, they found themselves answering an overly narrow question: does this impact user privacy? What makes them out of touch is that these were apparently the only criteria that needed to be met. As opposed to, say, "how would people feel if we took our software, which occupies a space in their workflow that demands a lot of trust, and auto-updated it with a marketing campaign for a TV show?" I mean, if they wanted to answer the question "how many of our users like Big Bird?", and they crafted a pop-up survey to be delivered through this mechanism that did not impact user privacy, would that get the green light? It's not just about respecting user privacy and safeguarding data, it's about adhering to expectations. The fact that there was apparently not a single human in the loop with the power to stop this from rolling out that recognized this is concerning, and the fact that there's no mention of it in this blog post is doubly so. What people want to hear is not just "we will safeguard your data", it's "we will never waste your time or introduce any kind of risk by pushing things like marketing campaigns into your trusted software."
- pdpi 9y ago
  Yes, absolutely. Hence my describing it as an overly narrow question. My point is precisely that, in going the SHIELD route, they selected for people who would ask that narrow question, because that's what they're usually concerned with, rather than the bigger questions they should've been asking.