On Tuesday, June 9, 2015 at 5:25:09 PM UTC+3, Gijs Kruitbosch wrote: > > Firefox has always been about empowering the user. You never dumb things > > down for them, you give them a choice. If there's an add-on we want, we > > find it or we create it. The add-ons should not be implemented into Firefox > > outright, because then you're removing that "choice" by providing it to > > people who never asked for it in the first place. > > We give people a choice, but we do make a "what's the default" choice, > no matter which feature or add-on is concerned. We pick defaults that we > think make sense. Implementing things that add-ons provide as default > doesn't remove choice (assuming things can be overridden or turned off, > like with pocket, and/or don't seriously interfere with a large number > of users' usecases (I don't think we need a built-in "off" switch for > bookmarks or tabs, for instance)). It changes the default behaviour. I > would contend that it is fully part of Firefox being Firefox (and indeed > probably any good product/browser) that it continues to try to have sane > and useful defaults.
1. There is a difference between blessing an option and making an option the default. Including Pocket by default would have meant preinstalling a Pocket addon. What has been done here is privileging Pocket over alternative solutions. Is is not the same as providing a default search engine. The only advantage a search engine gets by being provided by default is being the one available right after installation and nothing more. It can be replaced by the user at any time with any alternative and the behavior remains the same. On the other hand you can not replace Pocket. You can disable it and install an addon for an alternative but Pocket will still be there (not loaded in memory, but still there). You can point browser.pocket.api somewhere else but alternatives would have to follow the Pocket API. But things are not equal. There is nothing to say that after a radical update like FF28 ->FF29, Pocket wouldn't magically be enabled back.In doing so, you are endorsing Pocket more than just offering it as a default option would have. This is made worse by points 2 and 3. 2. Pocket is a Software as a Service provided by a for-profit company. It is SaaS and by definition the user can not control it. It is not freedom respecting because: A) there is no self-hostable FLOSS Pocket Server available for anyone to use instead and B) Read it Later Inc. controls the API. (that comment about non-free Javascript was an attempt to derail the discussion) 3. To make the above points worse, Read it Later Inc. gets access to private data (email address, reading list, timestamps, etc). If this was done using a locally encrypted file and synchronized using a blessed (as in point 1) file hosting SaaS provider (let's say Dropbox), the situation would not have been as bad because the list itself would have been protected by a layer of encryption and only the user would have access. The SaaS provider would only get timestamps and maybe an email address. As someone above already said, endorsing a non-free SaaS solution by blessing it and in doing so encouraging users to give private information to a for-profit company goes against the Mozilla Manifesto. There is more to say about Mozilla betraying users trust with previous moves but that can get off topic quickly. _______________________________________________ governance mailing list governance@lists.mozilla.org https://lists.mozilla.org/listinfo/governance
