I think you're misleading in quoting a two-year old blog, then closing off with "ooops", implying that Mozilla was caught in a privacy-related oops.
When in fact, Mozilla was super duper clear in the blog about the privacy implications of this experiment. And in the past two years, they have been focusing more and more on the privacy angle.
At the moment, both Safari and Mozilla look to me to be leading in privacy.
Being super duper clear about having sent the full URL history of users to a third party doesn't excuse you from the fact that you, well, sent the full URL history of users to a third party.
And unless the entire leadership of the Mozilla Corporation has been replaced since then (a wet dream of mine) that blog entry is relevant, since they've done it and they could do it again.
Are you saying that nothing could ever excuse that fact?
Cliqz is building a European, independent privacy-oriented search engine. This seems a worthy reason to me, especially with the complete and total transparency here. And in any case, the experiment has ended.
I still maintain your comments are misleading, by leaving out the context.
Setting aside whether Cliqz could be trusted or not, just think of it: this is a decision that went through lots of managers in the Mozilla Corporation, and nobody ever stopped to think for a moment "sending the entire URL history of our users to a third party? well this is wrong". This is the kind of decision that should've led to lots of resignations inside the company, but nothing happened. To me this means there's something deeply rotten inside Mozilla. What kind of assurance do I have that they won't pull this off again in the future? And let's remember that they are strapped for cash, so it's not like they don't have an incentive to sell private user data again.
"nobody ever stopped to think" is used constantly, yet is almost never the truth. What is infinitely more likely is that many people raised eyebrows and were assured by someone way above their pay grade that "everything is fine". Calling Mozilla "deeply rotten" is an extreme leap that (imo) is not justified.
Chrome has continuous and proven censorship? I think they are more or less on the same page than Firefox: they have their curated "app store" and they inconvenience you if you try to bypass it.
I would use Firefox if they actually were more open and free than Chrome. But if they are the same, just slower, then why would I want to support Mozilla?
I live in Spain and I haven't been to many mcdonalds restaurants but I don't remember there being any music at all? Just loud chatter. But people here are generally loud so perhaps they simply didn't think it was a good idea to have music.
I don't know much about how they looked in the past but currently they have sober colours. This is what a current restaurant looks like: https://s3-media0.fl.yelpcdn.com/bphoto/RxMy6l6HjBZujf7pHjtR... Seems like mostly hamburger colours? Mustard, lettuce, tomato.
But they weren't lumped together. They are distinct line items in their post. Not sure what the concern is anyways, their claim is clearly supported by the numbers: games are much bigger than music and movies. You can ignore the mobile games figure and it's still true.
Somehow. Her grandfather was from Hongkong but she was born in Germany. As far as I know she has been living in Asia for a couple of years and studied in Oxford and at the MIT.
Lap soldered mains connection (can break off easily). Not double insulated (think small children and fingers). Also the actual ioniser port is usually recessed to avoid accidental contact as it hurts if you touch it.
Also lots of hand soldered MLCC caps are high potentials. They like to crack and explode if you do that. They need to be machine soldered really.
Edit: also the design doesn’t use proper voltage rated resistors on the output. I haven’t checked the other parts but it’s probably marginal. You’ll get home one day and find your house burned down.
what about "manual" solder paste reflow ? I think I saw a few vids where people hinted at the fact that it's uneven heat and physical stress from handling that caused damage
Those needles connected to active make me uncomfortable, though there's a 10M resistor, I'd have an isolating transformer on the mains - not an engineer though.
Fuse would be good. Here in UK there’s one in the plug. The 10M resistor will current limit. Or probably flash over here.
Isolation transformer is not required if the enclosure is better. It also doesn’t help you if you stick yourself across it somewhere, only helping yourself with respect to ground. If you’re testing it or doing development with it I would use one. But I wouldn’t build it.
I used to hang around inside old valve radios when I was a kid. You learn respect for high voltages very quickly.
Hey you guys,
Author of the article here. Thanks for mentioning the issues. I never meant it to go this viral. Crashed my website in the morning.
You seem to mention that the output resistor was wrongly chosen. Can you elaborate? I thought I designed it properly.
This post was not ever meant to be used without using basic safety features in mind. (Like keep it away from children and all). I will explicitly mention that to be taken care of in the post.
Happy to take any more constructive criticisms to make the post better.
Yeah no worries. We’re all hyper critical. Need to outline the dangers because people who are unaware of the issues working with mains tend to be the first to drop dead. This is followed by the people who suffer from Dunning-Kruger effect. Us amateur radio operators like to kill ourselves on vintage power amplifiers doing that.
The output resistor need to handle potential to ground so say 7kv across output to ground. Resistor is rated for 200V working voltage. Usually you use a chain of many smaller resistors here.
I haven’t reviewed the design entirely but the main problem is the enclosure and mains connection and the component choice. The mains connection should be clamped in the enclosure and either spade connections or screw terminals or ferrules. The enclosure should be double insulated. As mentioned a fuse would be a good idea as well.
It's 20M, which if connected directly across 240V causes 12uA to flow --- that's in the range of the natural leakage current in most SMPS, and not harmful at all.
https://blog.mozilla.org/press-uk/2017/10/06/testing-cliqz-i...
"Users who receive a version of Firefox with Cliqz will have their browsing activity sent to Cliqz servers, including the URLs of pages they visit"
Ooops.