philipn

Me and my cofounder Chris have been quietly working on Encycla for a while now. Previously, I started LocalWiki[1] and DavisWiki[2]. What I couldn't stop thinking about was: is Wikipedia really "it" in terms of web-wide knowledge sharing?

There's been lots of note-taking apps and personal knowledge sharing tools developed in the past few years. But there's a big difference between working with people you already know and collaborating with anyone on the internet.

Right now, if you're interested in, say DIY air purifiers[3], you could throw up a document or webpage. But there's no good way for people you don't already know to work on it, to make it their own. If you're writing software, the answer is obvious: publish a Git repository on GitHub/GitLab.

With Encycla, we're building a sort of "GitHub for knowledge": a place where you can create simple, topical webpages that others can fork and asynchronously push & pull changes from (without knowing about Git or anything technical).

On the backend, every page on Encycla is a git repository containing Markdown that you can clone, edit independently of the Encycla website, push to other services (such as GitHub, GitLab), etc.

1. https://localwiki.org

2. https://localwiki.org/davis

3. https://encycla.com/Corsi-Rosenthal_Cube

Aaron Swartz actually built the original version of the Open Library site.

Yes, this is one reason why an approach using a form of soluble ACE2 is so promising; mutations that would normally cause e.g. monoclonal antibodies to stop working (as detailed with the mink mutation) would in theory not stop the 'decoy receptor' approach.

Despite the fact soluble ACE2 is not looking like it'll be developed in time to impact the pandemic, there is a silver lining:

If you look at the original paper (https://www.thelancet.com/journals/lanres/article/PIIS2213-2...), you'll note they include details on angiotensin peptide levels in the patient.

Interestingly, in this n=1 case report, it looks like the soluble ACE2 is unlikely to have worked as an antiviral; the patient already had a high level of antibodies, as you can see in the Lancet paper above. So /if/ the patient's improvements can be ascribed to the soluble ACE2 treatment, it's likely the improvement was because of the effect the soluble ACE2 had on the person's renin-angiotensin system.

And if this holds, then it's likely many of these same improvements would be possible using a common class of widely-available blood pressure medications called ARBs.

This approach would not neutralize the virus, but instead stop the body from damaging itself. That's the theory, at least.

There are ~20 trials of ARBs for COVID-19 in progress now, though almost all are underdosed.

The spike protein downregulates ACE2[1], which may make an otherwise benign respiratory infection severe[2].

1. SARS1 spike protein itself, without any virus, causes ACE2 downregulation which leads to severe lung damage: https://twitter.com/__philipn__/status/1237588716236898304

2. When mice are engineered to be deficient in ACE2, the otherwise non-severe RSV causes severe disease: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4728398/

(This may be why COVID is bad. Check out my twitter above for a lot on this)

It was not blinded. It was open-label and the outcomes were subjective (ICU admission). The doctors involved with the trial knew who got what and may have biased their actions based on that. Given the subjective outcome and small number of participants, this is particularly important. The trial would be slightly underpowered even if it were blinded.

Other studies on Vitamin D are retrospective cohort / observational studies which are heavily biased with possible confounders with respect to Vitamin D.

My personal opinion is that Vitamin D may help in COVID-19 but this trial does not 'confirm' that. Someone should run a similar trial with more participants and/or a fully blinded trial / a trial with more outcomes (biomarkers, etc).

I think the table in that paper is just confusing. The parenthesized numbers are for portion of the total number of people, rather than the portion of that category.

For instance, there are 137 current smokers. 108 are listed as non-severe. 29 are listed as severe. This means 108/137 = 79% non-severe, 29/137 21% severe.

Never smoked: 927 total, 793/927 non-severe (86%), 134/927 severe (14%).

Hope that makes sense. That table confused me too. Don't start smoking!

Please see my twitter account for lots on this. I'm in contact with a bunch of top researchers on this topic (renin-angiotensin system; ACE2/Ang 1-7/Mas axis; ARB usage for viral diseases) and will be posting a summary update soon:

https://twitter.com/__philipn__/status/1235756671852589056

It may be the other way around: ACEi/ARB may be protective. HTN without ACEi/ARB would be non-protective, potentially so much so that it skews the group fatality numbers. This is because people with HTN have a different renin-angiotensin system profile: more AT1R, less AT2R, more ACE, potentially less ACE2.

This is being looked at now. But the key is that in every study ever done on viral lung disease, etc, AT1R blockade was highly beneficial, and we know that ACE2-knockout basically screws up lungs, makes viral lung disease way worse, etc. Please see my twitter posts.

The virus eats up ACE2, downregulating it as it binds. This would have delirious effects. Check out the linked "essential reading" twitter post.

Will be posting a summary to my twitter soon.

List of repurposed drugs (not comprehensive) https://www.nature.com/articles/d41587-020-00003-1

Review of small molecule therapies (not comprehensive): https://blogs.sciencemag.org/pipeline/archives/2020/03/06/co...

It is related, but it's pretty complicated. For some information on this and the potential relationship, see my twitter account: https://twitter.com/__philipn__

Edit: contact made, thank you!

If anyone is reading this with connections to the Gates Foundation: Please contact Dr. Robert Kruse at John Hopkins. https://twitter.com/RobertLKruse. I have been trying to get in touch with people at Gates. This announcement unfortunately features no contact email address :(

Dr. Kruse is developing an extremely promising therapeutic, ACE2-fc, which will both neutralize the virus and treat the symptoms of the disease. It has been shown to work in vitro; variants have been tested in animals models; and its been through Phase II human trials for a different indication. A variant of the therapy, soluble ACE2, is being trialed in humans now in China.

Details on the approach can be found in his paper here:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7029759/ see "ACE2 immunoadhesin strategy"

Are registries allowed to charge different prices for different domain names?

E.g. can they ask google.com for $1B to renew and mygrandmascookiecompany.com for $20 to renew?

One reason is constitutionality. The "One man, one vote" principal likely has to be upheld for a voting system used in the US to be ruled as constitutional.

https://www.fairvote.org/how_ranked_choice_voting_survives_t...

https://www.fairvote.org/_ranked_choice_voting_reaffirmed_in...

Ranked Choice Voting is sometimes called Instant Runoff Voting when used in a single-winner election. But it can be used in multi-winner elections.

If it's used for a multi-winner election, it gives results that are proportional in representation. For instance, this is how it's used in the city council elections in Cambridge, MA. Sometimes it's referred to as "Single Transferable Vote" when used for multi-winner elections.

"Ranked Choice Voting" is an umbrella term that refers to the ranked ballots, round-by-round elimination of candidates, etc.

I loved this passage! --

"..There was zero overlap between the provers and the bulldogs. I was expecting at least some overlap: somebody who mocked me but also provided a valid solution, or even tried but failed. But that didn’t happen.

I normally assume these people are “brilliant jerks”: they’re assholes online, but I still have to listen to them in case they say something important. This really cracked that assumption: none of the “brilliant jerks” were willing to put any skin in the game. You don’t have to listen if they have nothing to say."

It's worth noting that this technique only works if the target user is using a mobile app (not web) that's been granted persistent location sharing permissions. The app also has to support one of the ad networks allowing the targeting they used in the paper. They tested this with the Talkaphone app, which requires (not requests) persistent location sharing.

I'm not sure, but iOS's anti-ad tracking function(s) may have an effect as well (https://support.apple.com/en-us/HT202074)

I believe a native mobile app would be able to make the request and scrap the data, regardless of CORS. Because basically no suspicious information is transmitted, it'd be pretty easy to squeeze past an app store review.

Of course, you could buy one of these services and have access, too.

At least in the current discussion, Linus says that eventually moving away from SHA1 in Git is a "no brainer," but that this announcement isn't a "sky is falling" moment. (e.g. https://public-inbox.org/git/CA+55aFz98r7NC_3BW_1HU9-0C-HcrF...)

You can check out the git developer mailing list discussion here: https://public-inbox.org/git/20170223164306.spg2avxzukkggrpb...

My rough summary: given there is no known second-preimage attack on SHA1, this is not an immediate danger to Git security because of the way Git works. The Git developers do want to move to a non-SHA1 hash at some point in the future.

Linus, from thread:

"I think that's a no-brainer, and we do want to have a path to eventually move towards SHA3-256 or whatever.

But I'm very definitely arguing that the current attack doesn't actually sound like it really even _matters_, because it should be so easy to mitigate against."

San Francisco has started the process of creating just that kind of end-to-end open source voting system (using the existing paper ballots, optical scanning and manual recount process that exists here). Here's some more information: https://gcn.com/articles/2016/06/02/sf-open-source-voting.as...

FYI, this isn't actually git, this is something entirely different that also has `git` in the name.

The Sanders campaign very quickly withdrew their request: http://boingboing.net/2016/01/15/bernie-sanders-campaign-wit...

Thanks for sharing this!

Why not include the cost of your time to maintain and improve the service? Without including the cost of your own time it seems hard to make budget decisions (e.g. the tradeoff between a PaaS and using a provisioned VPS). Even if you're not paying yourself anything, it seems important to include what you could be making if you were doing something else with your time.

Random manual recounts are done on a statistically significant portion of paper ballots. Additionally, paper ballots allow for manual recounts in contested contests (e.g. a legal challenge).

E.g. see https://twitter.com/SFElections/status/534405668258594816 and https://twitter.com/SFElections/status/535541022915186688

Please see my comment below. The author of the piece mis-captioned the photo. This is not an electronic voting proposal. It's simply making the existing multi-step process that San Francisco uses open-source. For instance, we currently use optical scanning machines here (as do most cities). With this proposal, those machines will run open-source software.

The caption on the image ("If San Francisco moves forward in developing an open source voting system, paper ballots could be a thing of the past") is incorrect. The open-source voting plan has nothing to do with getting rid of paper ballots. Paper ballots will stick around.

San Francisco's election process has many distinct pieces, some of which currently involve software. For instance, the optical scanning machines that scan paper ballots run software. With this proposal, these machines will run open-source software.

The proposal isn't for a new electronic voting system (e.g. electronic voting machines), or to rid the city of paper ballots. It's just to improve the existing process.

I think what they mean is that the previous Standard map style, which is produced by Mapnik, was rendered in such a way that the rendering process didn't have access to all of the possible OSM tags. But in the future they can use Mapnik to render more tags than they have in the past because Mapnik has improved in some way.

This isn't (or ever really directly indented to be) any kind of generic mapping service, or a map-driven interface for Wikipedia. Wikimapia (and other services!) provide interfaces for that. This is the first draft of a tile service that may be integrated into different spots throughout the project -- not an alternative to those kinds of things. It's basically just because Wikipedia can't use a tile service like the default OSM mapnik server without totally crashing it, so they need to build their own tile rendering pipeline, etc.

In case you're wondering what this is about: the scale of Wikipedia makes it difficult to just drop in a third-party OSM tiling service, so they're rolling their own map rendering process using OSM data. This is a beta of the rendered tiles, displayed inside Leaflet.

If you're into maps and this kind of thing, I highly recommend checking out LocalWiki (https://localwiki.org) too!

Yes, almost all employee (and founder!) shares are vested over a period of time. Typically, shares are vested over 4 years, with a one year "cliff." This means if the employee (or founder) leaves before the one year has elapsed, they don't get any of their shares. Then, after this set "cliff" period has ended, they get some fraction of their total shares each month.

But based on your question, I'm guessing you haven't yet set up your own founder share structure yet? If so, that's probably the best thing to do first. Sites like Clerky.com can help a lot like this, and you can figure out how all the bits and pieces work together with a good amount of googling and persistence. Of course, if you have some money, talking to an attorney about this can really speed up the process and help make sure you don't make any obvious mistakes.

Whoa, AppCanary looks really sweet now! I'm also really happy they added a lot more information about themselves to their website :) I think there's a huge market for really simple security products like this, as most people don't realize how much work keeping even an LTS Ubuntu image up to date is (hint: unattended updates aren't automatically installed, and you'll often have to restart the box to get essential kernel security fixes!)