notanote

When using netboot, the boot rom will pull the files it needs from the tftp server. It will first look for them in a subdirectory with the Pi’s serial number as its name.

If you’re netbooting without any sdcard installed, I’d suggest disabling sd polling.

That’s been available on earlier versions of the Pi (2B v1.2, 3B, 3B+).

It was not available at launch for the Pi 4, but was released in beta firmware in late September, and is currently part of the critical firmware update.

It works well, allowing separate boot images for each individual Pi on your network.

Rhinovirus is another well known cause for a common cold. A quick search result: webmd [1] gives 10%-40% for rhinovirus, 20% for coronavirus. CDC [2] has rhinovirus as the most common cause, and adds: “ Other viruses that can cause colds include respiratory syncytial virus, human parainfluenza viruses, adenovirus, human coronaviruses, and human metapneumovirus.”

[1] https://www.webmd.com/cold-and-flu/cold-guide/common_cold_ca...

[2] https://www.cdc.gov/features/rhinoviruses/index.html

‘All of Bach’ is the Netherlands Bach Society project to record the entire catalogue of Bach works. Check them out if you’re in the mood for classical music.

https://www.bachvereniging.nl/en/allofbach

Tl;dr of a dutch article[1]:

In the Netherlands there are between 40-50 covid-19 patients in ICU in critical condition. Over half of them are younger than 50. (Youngest is 16yo)

Mortality is much higher for the older group, but hospitalisation happens a lot in the younger age group as well. Obesity is a high-risk factor.

A comment in the reddit post[2] which discusses that article notes that in China the 15-49 age group made up 41% of severe/critical cases.

(The below 50 age group accounts for a bigger part of the population to be sure. I see no data for hospital admission rate as fraction of the age group.)

[1] http://www.ad.nl/dossier-coronavirus/40-a-50-nederlandse-cor...

[2] https://old.reddit.com/r/Coronavirus/comments/fj1owh/over_ha...

This is a different variation of the double-slit experiment, but it made me think of this result from last year, which pushes the boundaries of QM effects steadily further into the classical world.

“2000 atoms in two places at once: A new record in quantum superposition“

https://phys.org/news/2019-09-atoms-quantum-superposition.ht...

“Can the wave function of an electron be divided and trapped?”

https://news.brown.edu/articles/2014/10/electron

“The real danger is not that computers will begin to think like men, but that men will begin to think like computers.“

— Sydney J. Harris As quoted, without citation, in Howard W. Eves, Return to Mathematical Circles, (1988), 63.

https://todayinsci.com/H/Harris_Sydney/HarrisSydney-Quotatio...

I’ve run a Pi 3 as a PPS clocksource. I used a netbooted aarch64 buildroot on ramdisk, experimenting with rt-linux as well.

I saw similar performance, including those occasional spikes. I found that the flood of interrupts caused by the usb chip - they are handled in the driver rather than on chip - were detrimental to accuracy. Temperature variations were another source of inaccuracy. I didn’t figure out how to connect an rtc together with the gps hat.

I might try again on a Pi 4, I think it’s possible to power off the usb chip without losing ethernet as it does on a Pi 3.

For a fun variation: Nordic yoghurt cultures work at room temperature. Viili (Finnish) and filmjölk (Swedish) are a bit more kefir like, and (fair warning) not to everyone’s taste. They are a combination of leuconostoc and lactococcus strains.

It might be harder to get a good starter for these in the US. Keeping a yoghurt culture in good health requires commitment too. It’s like having a pet, it needs regular maintenance.

To clarify, the European Commission had a public consultation on DST[1]. The response was strongly in favour of ending the clock changes, and furthermore favoured permanent DST. The EC has called on member states to come up with an appropriate way of accommodating the public’s preference, telling them to coordinate so that time differences across borders are kept to a sensible minimum.

No decision has been made yet to either abolish DST, or move the clock forward one hour permanently, and it might not happen at all. The member states are supposed to reply before next summer, so that the switch to DST next spring could be the last time changing clocks in Europe.

[1]http://europa.eu/rapid/press-release_IP-18-5302_en.htm

They can and do update minor versions, but not as quickly as arch. They said to expect 4.18 for F28 in two weeks time.

A 70% solution is considered the most effective. Higher concentrations of alcohol (over 90% especially) evaporate more quickly, reducing contact time. Also, they work too quickly. By coagulating the outer proteins of bacteria instantly, they form a protective layer that actually stops the alcohol from penetrating further.

The Netherlands, the public broadcaster NOS:

https://nos.nl/teletekst

What advanced settings allows is to edit the blocking rules with something like the uMatrix gui. Once advanced is set, open the uBlock dropdown, then click on "requests blocked".

You will see a table with three columns. The second column edits the global rules, the third one is for local rules. Each of the rows in them shows three bars once you hover over one. Green = allow, grey = noop, red = deny. (Noop is the setting to not overrule the adblocker lists.)

In the "all" row at the top, click on green in the global column, to allow everything everywhere. Click the lock icon. Now uBlock won't interfere at all.

On a site where you want to block according to the lists, open the dropdown, and in the row with the site's domain name, click the grey bar in the local column. Click the lock icon to save the setting across sessions. uBlock will apply the lists' rules to that site.

If this is unclear, I can make a guide with images.

Ublock Origin can work this way with a couple of clicks. Use the advanced user option. Allow all content globally, and then use local override on sites where you want ad blocking.

What I did for my iPad2 was to disable js on safari, which makes a big part of the web fast and usable again, in combination with reader mode. I had Brave istalled for using websites more interactively, and for sites like youtube. It has adblocking without any hardware requirements. If Brave still supports ios 9, you could try that. It will still lag, and certain apps were simply unusable for me (protonmail), but I increased its useful lifespan this way.

That seems a little slow. I compile nightly for my own use and it takes around 20 minutes on a haswell i5, using tmpfs for the build directory.

Yes, once you turn on the "I am an advanced user" setting, you can fine tune script blocking in Ublock Origin. It's called dynamic filtering.

https://github.com/gorhill/uBlock/wiki/Dynamic-filtering:-qu...

As far as I know, the over the counter 500mg dose is sold over here as a painkiller, it's certainly not meant for long term preventative use.

A mostly fluid filled skull is not an indication of brain death, in fact there are cases of people with most of their brains spread in a thin layer against their skull. [1]

The name for aware but apparently comatose cases is locked-in syndrome. [2] A nightmare scenario for sure.

[1] http://www.rifters.com/crawl/?p=6116

[2] https://en.m.wikipedia.org/wiki/Locked-in_syndrome

They have done so recently, mp4parse-rust is a replacement for the old mp4 metadata parser.

https://github.com/mozilla/mp4parse-rust

If I've understood correctly when I read about this before, programs that overwrite before unlinking do not work for an SSD. The SSD probably will not write the data to the same location as the old file. Instead, use fstrim to have the disk reclaim all free space. After that it's supposed to be impossible to recover.

https://en.m.wikipedia.org/wiki/Trim_(computing)

That is great, thanks. It just seemed odd to compare compilers of such different age. I have nothing against clang, it certainly is faster in most cases.

As others said, it makes sense for everyone to try different toolchains and see what works. It is possible to test newer gcc versions by installing in /opt or something, just make sure to get the right libstdc++ at runtime. With ABI changes, updating the distro gcc is too much trouble.

Am I reading this right? He says he is comparing clang 3.8, which is a month old, to gcc 4.8, which branched over 3 years ago. I would rather see what gcc 6 does with this. That release should happen in a month or so.