[libreoffice-design] Minutes from the UX/design meeting 2022-Jul-13

[John Mills]

I would suggest to concentrate on the requested benefit for LibreOffice users first and why this measure is  being proposed to train some perspective.
In your example you propose by some percentage will inadvertently select to opt in to the anonymous non mandatory data collection. And some amount of non personally identifiable user metrics will be collected and processed to improve the LibreOffice user experience.
Okay, that's a distinct possibility, people do make mistakes. However at which point does that  situation detract from the benefit to TDF of having this anonymous metrics and the people who genuinely wish to improve the LibreOffice user experience by legitimately allowing anonymous metrics to be collected? 
PolyPoly is a cooperative as explained in the meeting with their source code being dual licensed. What is your specific 'fear' of what will be collected for this group of people that mistakenly opt into this process? Do you have concerns that PolyPoly have nefarious motives and and will work against TDF and release purposely or sell metrics on open source Office office usage?
If you do believe this then what in your opinion is the better way to proceed to gain these valuable insights? At the moment your posts do little more than scaremongering and you make assertions that TDF proposing this as a revenue stream which from everything that has been explained is patently incorrect.
John




Sent from Yahoo Mail on Android 
 
  On Wed, 13 Jul 2022 at 11:56 pm, Eyal Rozenberg<eyalroz1 at gmx.com> wrote:   On 14/07/2022 0:44, John Mills wrote:
> Do you consider people manipulated by asking them an unambiguous
> question that would default to no?

"Would you like to place a bet of 100 EUR on a die roll, winning an
extra 100 on a 6 and losing otherwise? [N/Y]: N_

Suppose I ask a million people this question. Some non-negligible
percentage will say Yes. Some of them have misread; some have
misunderstood; maybe they didn't read at all and just assumed they
should say Y for LO installation questions. Maybe the actually
momentarily in a betting mood and I caught them in this moment of
weakness. But I've manipulated all of these people.

And that's with a question that's worded fairly; your question won't be.
How do I know this? Because you've basically indicated what phrasing you
expect to have:

> Also are LibreOffice users too
> 'stupid' to understand that they are empowering innovation by answering
> a request that would not collect and personaliy identifiable data?

I'm sure that's how you'll sell them onto accepting your data collection
+ money-making scheme. And many people will click "yes" without
realizing what this means.

> Why is it unreasonable to collect data with the express permission of
> users

In my example above, you will be collecting X% of people's 100 EUR after
they have given express permission. Plus, like I said, you will likely
manipulate people into giving their permission, so it won't be _that_
express. It will make LO another one of those shady applications that
you have to warn people about.

 > with the sole aim of improving the application and user experience?

But this is not _quite_ the sole aim, now is it? Or did polypoly stop
being a for-profit organization while I wasn't looking?


Look, even if someone might convince me some data collection is
appropriate (which is not impossible, but hasn't happened yet), your
rhetoric so far makes me feel it's probably for the best if polypoly
were not involved in it.

-- 
To unsubscribe e-mail to: design+unsubscribe at global.libreoffice.org
Problems? https://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: https://wiki.documentfoundation.org/Netiquette
List archive: https://listarchives.libreoffice.org/global/design/
Privacy Policy: https://www.documentfoundation.org/privacy
  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/libreoffice/attachments/20220714/5161bc25/attachment.htm>