4 ms·
> We need a business model for privacy, which means a business model for paid client-side software. Until people are willing to pay for a secure browser, Mozill
by hackuser 12y ago
> We need a business model for privacy, which means a business model for paid client-side software. Until people are willing to pay for a secure browser, Mozilla will remain dependent on search engine revenue, which in turn derives from advertising/surveillance.
A very good point. Isn't Mozilla working on that? See today's post on sponsored tiles on new tabs:
https://news.ycombinator.com/item?id=8237812 https://news.ycombinator.com/item?id=8237812
In earlier discussions regarding that feature (I don't know if it subsequently changed), Mozilla said they were developing a way for end-users to control their own tracking data (rather than have it controlled by a commercial vendor) and provide it as desired (and anonymize it?). If that solution could work successfully for Mozilla, it could be a model for one possible solution for everyone, and be proof of concept when legislators and others are looking for options that don't disrupt many large businesses.
- walterbell 12y agoAny revenue diversification is good, so let's hope this works. But this is trying to remove the search engine middleman by working directly with advertisers, i.e. Mozilla would still be in the advertising/surveillance business. Here's a better option: partner with a (micro?)payments company to maintain a secure client-side database with dashboard for donations to open-source software and CC-licensed content. Make Mozilla the best browser for consuming, discovering and funding OSS and CC-licensed content. Instead of showing "sponsored tiles", show project analytics for all the wonderful things that were done today by the projects to which I have donated. Make Mozilla the CRM user interface and payments broker for open content. This would have the side effect of funding new, open content. In this possible future, Mozilla becomes a funnel of money to the commons (including itself), not a shopping cart or magazine. This future would not conflict with sponsored tiles -- let the user choose. Edit: Mozilla could also modify the browser so it can perform "federated search" on the local device, trusted peers (via webrtc) or vertical search (e.g. HackerNews, StackOverflow, Wikipedia) before querying a public search engine. Enable a split-screen mode where private/federated search results are in one half of the screen, and public search results are in the other half of the screen. This will route traffic & revenue to community sites and will promote search privacy, i.e. always hit the local & community cache before querying the broader web.
- hackuser 12y agoVery interesting -- for my prior post, I had written another few paragraphs on the need for a public micropayments clearinghouse and how well it would support free-as-in-speech software and media, but then deleted them as veering too far off-topic. (I'm glad you posted yours.) Why limit the clearinghouse to free content and techy interests? Why couldn't other publishers use it, especially those whose products are too cheap per unit to charge for them one at a time such as newspaper articles? I see your ideas on that and on search (a very interesting idea) as ways Mozilla can leverage their very valuable asset, user trust. Essentially Firefox would morph into a client for many services. But we are getting very far ahead of reality ...
- walterbell 12y ago> Why limit the clearinghouse to free content and techy interests? Good point. There would be less need for donation analytics in that scenario. If it was expanded to "everyone", there would be many more stakeholders at the table, e.g. every payment company. It would be easier to prove the concepts in the underserved market of "free" software & content. > But we are getting very far ahead of reality ... It could be prototyped with a Firefox extension that bridges between the browser and a local (Py)QT WebKit app. Xapian does local search on Linux.
- patcon 12y agoIn regards to giving data control to users, MIT media lab is worling on the openPDS (personal data store) system: http://openpds.media.mit.edu/ http://openpds.media.mit.edu/ Honestly kinda surprised mozilla doesnt have their hands on that as a partner, but hey, they can't do everything :)
- walterbell 12y agoOpenPDS is the reference implementation for http://www.whitehouse.gov/sites/default/files/rss_viewer/NSTICstrategy_041511.pdf http://www.whitehouse.gov/sites/default/files/rss_viewer/NST...