mic47

1350$ per month is still fraction of salary of developer. So if it saves you that fraction of your (or your colleagues) time, it is definitely worth it.

The automation they described sounds like automation of part of project management. why they are not selling that? :-D

well, maybe because in some other EU countries it's 22 EUR for 500Mbit :-)

Would this include ad-tech, like advertising cookies and other tracking stuff (specifically, not Facebook / google ones)?

The presentation actually talks about FB needed to solve challenges of training and evaluating the AI without getting the data to it's server, and I haven't noticed any point where whey would suggest that they need to break encryption, or get the unencrypted data to the servers.

If everything happens on the device and noting is sent to servers, how does it break encryption?

In case I am wrong, and they actually suggest that they will really send data back to its servers, where it is suggested? I really failed to find it.

Does it apply only to web browsers? What about apps? As far as I understand, huge part of Facebook, Twitter users use the apps and not the web, so if the bill does not ban it also there, then it will be kind of toothless (and motivate those companies to even more aggressively nudge people to use apps).

We had this in Slovakia (https://piano.io/ ), but it was not very successful, and all media companies now have their own systems. But they seems to expand to other markets.

I think that their main problem was that at the same time, the subscription was low (so didn't bring much revenue) and at the same time people hated it because now they had to pay for something that was free before. So I guess once we get used to pay for journalism, then something like netflix for media might work.

They are. It's just in today's terms, "smartphone == has touchscreen" and "feature phone == looks like old fashioned phone with keyboard".

Is it just me, or do "secret" have negative connotation? Because for me it does, and I am really sad from journalist when they use "secret" or "secretly" in headlines, when it just mean internal, or just not disclosed. What is interesting that the article does not contain word secret, and they always use the word internal.

yes, they are. For operation thing, console > web interface >> gui applications

Thou, there is 1 big advantage of web interfaces and that you can send a link to it to other person (that can make it better than console for some things).

> "problem of academics being incentivized to publish in the most prestigious journal possible."

Is this a problem? If you have interesting, ground-breaking, or just really surprising results, what is wrong with trying more prestigious journal? Problem is not that there are journals that take only "interesting" results, but that a lot of research is not published at all, just because results are as expected. And this journal fixes that.

> It would be far better if the most prestigious journals pre-committed to publishing studies based on their methodology and hypothesis, before seeing their results. This is quite interesting proposal, could work well for experimental sciences, but worse for stuff like math / computer science, theoretical physics.

Actually, your question and followup answer is valuable addition to the discussion.

To file patent, you have to have some prototype, but it actually cannot be in production for more than year (and does not have to be in production at all).

This works as contact matching: they provide list of contacts and FB show to those people ads, so advertiser will not gain access to your personal data.

I don't think you can target people by names or dates of birth at Facebook. IIRC targeting by name (even indirectly) is prohibited by Facebook.

Could virus mutate to be resistant to that? Sure, why not. Will it be super-virus? Depends on what is super-virus. If it means more deadly, then probably not, since this resistance is most likely not related to deadliness.

It is much easier to measure amount of scam than detect. That means that company can report "cleaned up numbers" (at least with error bars) but not able distinguish all bad stuff from good stuff.

Imagine this happening in the tunnel

So data portability is now bad, because Facebook is suggesting it.

true, apart from I would expect this to go from growth team (making signup as friction less as possible), not from anti-fraud team.

Just type Target or other company into the search field and you'll see that it looks like all ads: https://www.facebook.com/ads/library/?active_status=all&ad_t...

I guess you are right and the political ads are mentioned for emphasis.

This does not help at all. If you hash client side,then the hash will became your password. If you accidentally log this hash, anyone who have access to that hash can log in as you easily by constructing logun request.

What I said already assumes that you delete the all instances of that content when you find out it's bad. FB already does this. Problem is, that people see content before it is taken down. And once it's taken down, bad actors stop uploading it, and slightly change it (enough to make it pass existing upload filter).

Not really. There are 2 factors that play "against" FB having nice numbers.

It takes some time between unique (or in some similarity cluster or whatever they have) content is uploaded, reported and taken down. During this time, exactly same content is not being taken down, because it's not known to be bad. So even if all the content was byte to byte identical, if 20% of it is uploaded before it is reported, there is no chance to have >95% block rate automatically.

Once content is known to be bad, people (and bad actors) will usually stop uploading it (because it no longer works). Also people stop sharing it, because it's no longer possible (and FB delete all instances of such video from the platform anyways). If there is lame script kiddie that does not handle errors, then the number could be 95%, because it would continue with upload attempts. But normal people, or sophisticated actors will respond to errors and stop uploading exactly that video.

1 extreme example: with 1 sophisticated bad actor uploading exactly same video (through some fake/hacked accounts/pages, or whatever), the block rate will always look bad for Facebook: bad actor uploads N videos before being blocked. Once blocked (on video N+1), bad actor will stop. The block rate will be here 1/(N+1), which will be way less 80% (FB's current case) for any N > 0.

What did I try to say? Percentage of what is being blocked does not make sense, since it also depend on how fast is attacker detecting that is being blocked. If fast, the number will always look bad. If attacker does not give a shit, the number will look awesome.

Mostly, because article is mashup of random data things about Facebook, and it's not clear what is being investigated. They started that phone makers data deals are being investigated (which is basically alternative client), then goes into Cambridge Analytica, and mention Bing data access in the end (which could be about deep access you are talking about, but it does not looks like it's being investigated).

And for phone deals, criticizing friends/deep data access... If you want to build alternative FB client, your phone have to have access not just to your data, but also to your fiends data (i.e. all data you have access), in order to provide any useful service. So yes, technically your friends didn't agree that your alternative FB app can access to their data, but you gave your access credentials (and permissions) to access that data to your phone.

most startups can use their standard comm channels since most startups are not doing collaboration/communication tools.

Not all samsung devices had android at that time.

Yeah, I was kind of writing that under assumption "if it would have been found criminal".

Ok, so those partnerships were basically allowing Samsung and others to build Facebook app on their phones (i.e. allowing alternative client).

Would this means that it will be criminal to allow companies to create alternative clients? That is really interesting development.

Why? When Facebook is down, you cannot use Facebook to communicate about the issue. That is also reason why FB still uses* irc instead of messenger for coordinating how to resolve such issues.

* Or at least used 1 year ago, when I was working there.