arctux

Related earlier post:

I want to respond to the misleading media coverage of messages I posted about Marvin Minsky's association with Jeffrey Epstein. The coverage totally mischaracterised my statements.

Headlines say that I defended Epstein. Nothing could be further from the truth. I've called him a "serial rapist", and said he deserved to be imprisoned. But many people now believe I defended him — and other inaccurate claims — and feel a real hurt because of what they believe I said.

I'm sorry for that hurt. I wish I could have prevented the misunderstanding.

Source: https://stallman.org/archives/2019-jul-oct.html#14_September...

All these links are year or two old.

Two of them are more than a year old, but the practices described in each are ongoing. The third, which describes Google's tracking of users after they've specifically opted not to be tracked is from nine months ago.

A big push towards openness and privacy has happened over the last year.

After literally a decade of constructing what is very likely the largest database of personal information in the world. Since the late 2000s, when Google purchased DoubleClick, it has worked to collect information without the informed consent of its users. What fraction of your users know that Google purchases their credit card transaction histories?

What is the "big push"? The only things I can think of were the opt-in auto-deletion of a subset of data announced over the last week or two. All the user has to do is pay attention to the tech press, then remember to activate the feature when it launches at an unspecified future date!

What is this "openness"? Working on a censored search engine for China without informing their own head of security?

...people will still opt into the surveillance society willingly: because the utility it brings them outweighs other considerations.

Sure, they absolutely do. There can be significant utility gains from large collections of information. But much of the utility could be gained from information collected in a anonymity-protecting matter. In order to have traffic information, for example, Google doesn't need to continuously track your location history.

Ask people if they want to be tracked at all times, and they'll say "no". Ask people if they want to be able to locate their phone when they lose it, and their answer might be different.

And neither of these require surveillance. The phone could be located either by returning its location on command, or by uploading encrypted location data which only the user has the key to. Whatsapp, for example, shows that end-to-end encryption can be seamlessly integrated.

Ask them if they'd want be able to cal 911 and ask to come and help them even if they aren't sure where they are, and you'll get a different distribution of answers again. > In the latter case, lack of "surveillance" is seen as a "tragic shortfall" [0], and adding it is a "feature"[1].

Once again, this does not require ubiquitous surveillance, and it is misleading, at best, to imply that it does. Do you really not see the difference between location data provided to assist emergency response from a 911 caller and continuous location monitoring so that Google can serve more profitable ads?

And I can't vouch for all of Google, but regarding location data, Google has been pretty transparent regarding which data is collected and stored; papers like NYT covered it extensively - see [1].

How did you read that article and come away with the conclusion that Google has been "pretty transparent". The story was written after more than a year of other news outlets reporting on law enforcement using Google's location data to fish for suspects. Google has been providing this data for at least two years before the Times reported on it [0].

And moreover, Google has been consistently on track to store less private data.

Such as credit card transaction data collected without most people's knowledge [1] or location data after you've explicitly told it not to [2]?

Technology companies need to understand that both words "informed consent" are important. We currently have very little in the way of choices when it comes to data collection. It is simply not possible to opt-out anymore without tremendous effort and personal cost. I like this quote from Maciej Ceglowski:

"A characteristic of this new world of ambient surveillance is that we cannot opt out of it, any more than we might opt out of automobile culture by refusing to drive. However sincere our commitment to walking, the world around us would still be a world built for cars. We would still have to contend with roads, traffic jams, air pollution, and run the risk of being hit by a bus. Similarly, while it is possible in principle to throw one’s laptop into the sea and renounce all technology, it is no longer be possible to opt out of a surveillance society."

[0] https://www.wral.com/Raleigh-police-search-google-location-h...

[1] https://www.cnbc.com/2017/05/24/google-can-now-track-your-of...

[2] https://www.apnews.com/828aefab64d4411bac257a07c1af0ecb

I'm syncing around 90 GB between my server, laptop, and LineageOS Pixel phone. I use it to sync my documents, music, passwords, and archived pictures. I also use it to sync photos taken by my phone camera as they are taken.

Setup:

* Camera: 1.8 GB, 243 files

* Documents: 10.8 GB, 4604 files

* Music: 61.5 GB, 25077 files

* Passwords: 660 KB, 726 files

* Pictures: 16.5 GB, 6450 files

The passwords are managed by 'pass' [0], which is viewable on my phone using Password Store [1]. Cold-launching Syncthing takes ~10 seconds on my phone, but it does it automatically on boot and thereafter runs in the background. Battery impact seems to be negligible.

[0] https://www.passwordstore.org/

[1] https://f-droid.org/en/packages/com.zeapo.pwdstore/ and https://play.google.com/store/apps/details?id=com.zeapo.pwds...

I've only seen articles that say Google was going to "explore" adding differential privacy to Gboard analytics [0]. Do you know if the feature ever shipped, and is it the only way sends data to Google?

I'm mistrustful of Google's privacy stance, since they have a history of changing their privacy policy, then misleading users about it. Remember when they implemented personally-identifiable web tracking and sold it to users as "new features for your Google account"? Merging Doubleclick's tracking data with my Google account doesn't seem like a feature to me.

[0] https://venturebeat.com/2017/04/06/following-apple-google-te...

[1] https://www.propublica.org/article/google-has-quietly-droppe...

No. Fines exist to dissuade illegal behavior. The loopholes that Google (and Apple, and many other multinationals) should be abolished, and any tax evasion should be prosecuted. Antitrust fines do not exist to punish tax avoidance. They exist to penalize companies who use their size to prevent competition and distort markets.

I have many issues with Google's business practices, but punishing them is not worth throwing out the rule of law.

We also shouldn't pretend that the EU justification for fines is purely about citizen protection, they see this as a "tax" as well (though that is of course not their primary motivation). Otherwise, they would work harder on non-punitive approaches towards combating the ills of Google and the like. One would hope education and encouragement of alternatives and other positive-leaning approaches might be prioritized over a gavel.

The fine exists to remove the profit gained from the illegal behavior, otherwise performing the illegal actions would have a positive expected value. The company is also required to provide a list of remedies to ensure that the behavior is not repeated.

As for "non-punitive" approaches, the encouragement of alternatives would have to be non-financial since state aid is barred under EU rules [0]. I suppose they could recommend alternatives, but if it were really that simple to avoid a consolidation of power in the tech industry we wouldn't have such monopoly/duopoly issues.

[0] https://en.wikipedia.org/wiki/State_aid_(European_Union)

I think that this is a fairly common when it comes to technology. The terms and conditions seem reasonable ("we collect some data to provide more relevant ads"), but when you look a bit more closely they build a personal file that contains who you communicate (email/text/call) with and how often, where you go, what you buy, which websites you visit, which videos you watch, etc to the extent that they are able. My mother is very smart, well-educated (she has a PhD), and relatively tech savvy (she works in scientific computing), but she was still floored when I told her about some of the tracking Facebook and Google perform. Google recording her location (which she technically agreed to, but did not realize) was enough that she asked me to help her migrate away from Gmail. She probably would have managed without my assistance, but the barrier would have been much higher.

I'm emphatically not defending them here, but it may make sense from their perspective. They live and die by ad targeting and maintaining their monopoly power over social networks, both which requires large amounts of personal data. Knowing what apps teenagers have installed and how often those apps are used is instrumental in detecting an up and coming social media rival.

Further, their only major competitor in the ads space is Google, which has access to this information via Android and its control over the Play Store.

Plus, what are the teenagers going to do about it? Facebook also owns Instagram. I guess they could use Snapchat...

You seem to be correct.

Google Maps Platform is a part of Google Cloud, but not a part of Google Cloud Platform

I have never felt that I have been confused by something so justifiably.

This is not correct: https://cloud.google.com/maps-platform/

The Google Maps API is a good example. They increased prices by an order of magnitude while simultaneously decreasing the free tier with one month's notice. [0]

[0] http://geoawesomeness.com/developers-up-in-arms-over-google-...

I strongly dislike that they've done that. In their partial defense, the selection of recommended articles based on your browsing history is done on device.

Yes, but it's not the default. From the article:

One could, however, add a second passphrase that is never sent to the server, and encrypt the data using that. Chrome provides this as a non-default option.

The average user doesn't have the expertise to know that they have to configure an additional "master password" to keep Google from mining their data for ads.

This is a good example of why Firefox is so important. Mozilla's incentives, unlike those of companies making significant revenue from tracking-based advertising, align with the user. Google, for example, could have implemented Chrome's sync feature in a privacy preserving manner, but instead chose to use it as a method to collect their users' complete browsing histories.

My undergraduate advisor had a small hole in his retina due to a high-power laser. The injury would have been far worse if it weren't a visible laser, since his blink reflex greatly reduced the damage.

Curious, it appears that you guys use a highly focused beam on small particles, like 1 W focused into a diffraction limited spot, right?

That's correct.

How come the bio-molecules that you're manipulating with this light don't just "burn up"? Is it because they're mostly transparent at that wavelength? Or you're only exciting a marker molecule that is stuck to them?

We trap and manipulate a micron-sized polystyrene sphere that the proteins are attached to, not the proteins themselves. The microspheres don't burn up because they don't strongly absorb (i.e., they're mostly transparent) at the laser's frequency. Paper, on the other hand, absorbs strongly and ignites.

Optical traps rely on the momentum of light and the fact that a microsphere displaced from the diffraction-limited spot refracts light in direction of the displacement. Since momentum is conserved, the light directed away from the center of the trap creates a force that pushes the sphere toward the trap. If the light were mostly absorbed or scattered, this force would be along the axis of the beam.

Powerful lasers can be scary, especially if they're in the infrared. The first warning you receive that it's reflecting into your eye is blood seeping into the vitreous humor from your retina's capillaries.

Chu's prize was for laser cooling, which uses a magneto-optical trap. Magneto-optical traps confine atoms by exploiting the Zeeman effect, a spatially-varying magnetic field, and a laser with a precisely-controlled wavelength tuned to be slightly red of an electronic transition. Optical traps, on the other hand, rely on the refraction of a much larger, but still microscopic, object.

Some of Chu's work uses an optical trap, but that work did not earn his prize.

Yes, but high-power lasers need to be treated with respect. Reflections are capable of instantly and irreversibly blinding. There's an overused joke in labs that work with them: "do not look into laser with remaining eye".

If you're going to attempt to build one, please get a pair of laser glasses that are rated for the wavelength you are using.

Our strongest trap is powered by 1.5 W 1064 nm laser beam. (For comparison, laser pointers are usually under 0.005 W.) Unlike most other light sources, laser power measures the output power rather than electrical input power.

The beam will instantly ignite paper, and will cause damage to the most common type of laboratory laser beam blocks, which are made out of stainless steel.

We don't take the air out the system. Air is basically completely transparent in the visible and infrared. We sometimes flood a portion of the instrument with helium, since its index of refraction is ten-fold closer to 1 than air's. This causes the beam to fluctuate less due to air movement.

Kind of? It's the only way that I know of for a laser to move a particle towards the laser's source. But it has some severe limitations: the force is extremely weak, on the order of 0.1 pN/(nm W); the particles have to refract, rather than absorb or scatter, the light; particles larger than 10 microns in diameter tend to be melted or vaporized since they can't dissipate the energy quickly enough.

I want to emphasize how incredibly weak these traps are. In order to apply a maximum force of 200 pN orthogonally to the path of the laser to a 1 micron polystyrene sphere requires approximately 1 W of laser power. Further, it takes about five times more laser power to apply the same amount of force axially (i.e., along the path of the beam) compared to transversely (i.e., in the plane orthogonal to the beam).

I'm a grad student in a lab that specializes in using optical tweezers for single-molecule biophysics research. I'm out sick today with nothing better to do, so feel free to ask me anything.

Usually optical traps use 1064 nm NdYAG lasers, which is in the infrared. Most objectives aren't designed for infrared light, so the objectives used for optical trapping are usually customized.

Do you know if Mapbox has any plans to work with the World Bank's OpenTraffic[0]?

[0] http://opentraffic.io

You should try DNS66 [0], which does not require root. It acts as a VPN on Android and returns 0.0.0.0 when any app tries to resolve the IP address of an ad-serving domain. Other network traffic passes through unmodified, so it's very battery friendly.

The app is also available from Fdroid [1]. It isn't available from the Play Store, since Google has banned ad blockers.

[0] https://github.com/julian-klode/dns66

[1] https://f-droid.org/en/packages/org.jak_linux.dns66/

I was using DNS66, much is user friendly and very effective. Unfortunately, it is implemented as a VPN, and Android will allow one VPN to be active at a time.

I'm using the user space implementation.

I've been using Wireguard on both my laptop and my Android phone for about two months. I've been using the wg-quick systemd units, and everything has worked amazingly well. The only downside I've notice is slightly increased battery consumption on my phone, but that's to be expected (it uses approximately 5% of the battery per day).

I use dnsmasq to resolve DNS queries on the server side. Dnsmasq's configuration file includes entries to block connections to ad networks, based on Steven Black's host file [0]. It's a great to achieve ad blocking on Android, since Google has banned ad blockers from the Play Store.

Jason Donenfeld, Wireguard's author, has a Patreon page: https://www.patreon.com/zx2c4

[0] https://github.com/StevenBlack/hosts

It seems to be somewhat particular about relation IDs. I have it working for Palo Alto, with relation ID 1544955. If you want to try this for your city, I recommend zooming in the city on openstreetmap.org, clicking the "Query features" button on the right side of the map (it's the cursor with a question mark icon), and clicking within the boundaries of you city. On the left side, a list of features will appear. Scroll down to enclosing features, and click on the appropriate entry. A relation name and number will be displayed at the top, with the OSM data below.

I hope this stays within the Overpass API server's usage limits. Processing Palo Alto caused the page to download ~85 MB from overpass-api.de, which asks users to stay below 10k queries and 5 GB per day. The OSM wiki lists other servers, some of which allow heavier usage.[0]

[0] https://wiki.openstreetmap.org/wiki/Overpass_API