We are committed to allowing the maximum range of expression as possible for our users, and we have the history to back that up -- early on in our history, we went without any income at all for four months when PayPal closed our account with them after our refusing to censor content that was legal but unpalatable while we worked to find a payment processor that would more closely match our values around user expression. But every site on the internet that accepts user-generated content needs to set boundaries about what can and can't be posted on the site; to do otherwise would make a site unusable for you and create enormous legal liabilities for us.
The question of where to draw the line with content policy is a difficult one -- I've been working in this space for 20 years and I still don't have many more answers than I did when I started -- but there are categories of speech that are legal (or, at the very least, on the murky line of "possibly legal, possibly not") but severely harmful or damaging. We don't let people post other people's Social Security numbers, for instance, even though doing so isn't presumptively illegal, because of the severe offline harm it can cause.
Those questions are hard, complicated, and require a great deal of careful thought and balancing competing interests, not in the least because every specific case is going to differ slightly from the last, often in ways that aren't immediately visible to people who are looking at it from the outside or without context. No site ever gets them 100% correct, and a site should (and we do) regularly re-visit their answers to make sure those answers still align with their values and with the ways their site is being used. We believe that it's more important, not less important, for a site or service provider that commits to maximal expression to regularly review their policies to make sure that they aren't prohibiting too much or too little content.
Cloudflare's statement from yesterday indicates that they don't believe they should have any restrictions at all for sites that use their security services and that they regret past instances in which they denied service to sites that were causing tangible offline harm, distributing child sex abuse material, and inciting white supremacist terrorism. Precisely as a service that commits to maximal expression, we believe that attitude is dangerous and harmful, and we don't believe we can ethically continue to do business with a company that refuses to engage in even the smallest fraction of the amount of effort we put in, on an ongoing basis, to balance those multiple competing interests in ways that allow for maximum user expression while seeking to minimize concrete offline harm.
no subject
The question of where to draw the line with content policy is a difficult one -- I've been working in this space for 20 years and I still don't have many more answers than I did when I started -- but there are categories of speech that are legal (or, at the very least, on the murky line of "possibly legal, possibly not") but severely harmful or damaging. We don't let people post other people's Social Security numbers, for instance, even though doing so isn't presumptively illegal, because of the severe offline harm it can cause.
Those questions are hard, complicated, and require a great deal of careful thought and balancing competing interests, not in the least because every specific case is going to differ slightly from the last, often in ways that aren't immediately visible to people who are looking at it from the outside or without context. No site ever gets them 100% correct, and a site should (and we do) regularly re-visit their answers to make sure those answers still align with their values and with the ways their site is being used. We believe that it's more important, not less important, for a site or service provider that commits to maximal expression to regularly review their policies to make sure that they aren't prohibiting too much or too little content.
Cloudflare's statement from yesterday indicates that they don't believe they should have any restrictions at all for sites that use their security services and that they regret past instances in which they denied service to sites that were causing tangible offline harm, distributing child sex abuse material, and inciting white supremacist terrorism. Precisely as a service that commits to maximal expression, we believe that attitude is dangerous and harmful, and we don't believe we can ethically continue to do business with a company that refuses to engage in even the smallest fraction of the amount of effort we put in, on an ongoing basis, to balance those multiple competing interests in ways that allow for maximum user expression while seeking to minimize concrete offline harm.