Sry Moxie, you lost my trust when you temporarily took the signal codebase closed source to integrate some cryptocurrency scams into it. Now your criticism just sounds like whining about a more successful competitor. Why is Telegram winning against Signal ? Mainly because the app is more stable and doesn't dictate its philosophy to the user. Proper e2e encryption has tradeoffs. Why shouldn't the user decide for which conversations he is willing to place convenience over security ? Why does Signal force users to use a pin even if they don't want to ? Why do you lose your history when you switch between android and ios ? why would I choose a messenger that requires a phone number over another that does not ? I was an enthusiastic evangelist for signal. But over the years you just lost me. I still use it but I don't recommend it anymore. I know, I'm not the only one. Apparently you seem to think, you can make up for it with paid advertising that you plastered my city with (repeatedly). I would have preferred to see that budget allotted to developing a more stable client and new features, but I'm not calling the shots.
I used Google Authenticator some years ago on a phone that broke. At the time I was unable to restore it to my new phone. The proposed solution used to be to authenticate multiple devices (phones with Authenticator, Yubikeys) but to this day many Services only allow You to link a single second factor to your Account, meaning that upon loosing that factor you’re required to contact support to restore access to your Account. Is there any open source solution that generates a second factor in software and can be easily cloned/restored to multiple devices ? Nobody buys a single set of keys for their front door or car. I know why yubikeys don’t offer „spares“ but usecases can differ are there any yubikey protocol compatible competitors that offer spare keys ?
First to get this out of the way: I fully sympathize and hope some hn user with privileges at facebook can escalate this to the appropriate people to quickly get it resolved.
This isn’t the first post of this kind. What is new however is, that non-techies are signing up for hn accounts to get things resolved. I don’t blame CrzyLngPwd for advising her to post here, I probably would be giving the same advice if a relative was involved.
People working at FAANG‘s don’t log in to hn to fix their companies processes by performing ad-hoc tech support and they shouldn’t have to. We should all (whether FAANG or not) lobby management within our organizations to ensure that proper appeals processes are in place. It is in the interest of the employees as well as the company. Because while it was fine to have no appeals available when „The Internet“ was a fun hobby for a select few, it is decidedly not fine, when millions of livelihoods depend on the whims of automated appeals processes. You can bet your RSUs that governments will step-in sooner or later and institute tedious (and perhaps inappropriate) procedures set in (regulatory) stone, if the industry doesn’t manage to self regulate in an effective way. You don’t know when it’s coming, maybe the golf buddy of an influential Senator gets banned on Facebook without any recourse or the Wife of a Member of the European Parliament loses access to her Google Account and we might see this issue regulated out of existence. I‘d much rather the industry takes initiative now and solves this Problem on it‘s own. It costs some money but it’s not Rocket Science | Brain Surgery.
I‘m sorry for You. I can see how you‘d think that your software is providing society a service. And yet from my perspective your software is malware. I don’t think you have considered the legal ramifications of your software in all markets where the app store is offered. Aside from the legal risk to You, there is also a reputational risk to Apple, a company trying to rebrand itself by focussing on privacy (CSAM nonwithstanding). Do you have safeguards that can 100% rule out the use of your App as spyware ? I think you should move on with your life.
There is quite a lot of land within the arctic circle. If it continues to warm up it will become more conductive to human settlements and agriculture. hn is a place where most people see the upsides of change, I don’t understand why so few here admit the benefits of climate change.
Not only are they forcing the unvaccinated to get the shot, they are also forcing everyone who is fully vaccinated to get booster shots every 6 months for the indefinite future. When the vaccines were first introduced their validity was fixed at 12 months and their was talk about extending it ("because they are so incredibly effective").
That is my understanding from german cartel law (very similar)as well. However how do you structure a deal to take retroactive actions by regulators into account. I don’t know if there is a set of established best practices there, but I‘m no M&A Attorney so this could well be a solved problem.
As much as the Law can bore me at times there frequently are questions such as this that pique my interest again.
Saw your recent (6mo ago) ask HN. Did you end up changing careers or stick with the law ? I‘m in a similar situation.
How does such a retroactive block work ? I imagine Meta executives have already taken over the reigns and directed the company towards full integration with facebooks other services. If Giphy had had a project in the works that could disrupt any of Facebooks monopolies than surely that project would be canceled by now and anyone involved would have transferred internally to the relevant Meta subgroup or left.
So if blocking the deal results in Meta loosing control of Giphy but getting it’s money back, wouldn’t that be a great outcome for Meta as they would have achieved their main goal for free ?
The instance failed at mere DOZENS of simultaneous users. Clients kept retrying and then there were thousands of requests of users who were unable to interact with the instance. I went though the logs in detail. It was a clusterfuck.
Wow, this is beyond ridiculous. This is the NGO I was involved with. It did not scale to 30,000 it failed miserably at merely dozens of simultaneous Users. The linked Video is the Exec who took responsibility for ordering Polis trying his best at PR Crisis Management. He was fired shortly afterwards. Apparently You don’t understand German, because in the video he is admitting the technical problems with polis. I‘m just amazed how anyone would point to that video, to tout the supposed strengths of their platform.
The question is why your consulting services are necessary in the first place. Most Open Source Projects I know can be set up by anyone who is competent in the underlying technology. You explicitly warn people that they will be unable to set it up on their own and that they will need your services. You had years to simplify deployment and haven’t done so. Your Github is still lagging the version that you’re privately deploying.
Can You point to logs of any recent deployment that scaled to 100s or 1000s of simultaneous users ?
Somewhere else in this thread you claim that OSS software has a competitive advantage in the EU. Is this the reason you pretend that your software is OSS while in practice it is not ?
In 2018 I was involved with an NGO that planned on using Polis in their work. They contacted the maintainers who recommended contracting an agency which had experience with Polis as many of the core developers worked there. The NGO did just that, and paid a huge sum for the deployment. The first discussion resulted in a total overload of their servers. The NGOs marketing spent for this project was completely wasted and its reputation tarnished. I was brought in in the aftermath to communicate with the Agency and to scale the deployment. They were unable to scale even to 100s of simultaneous users. We tried to deploy it on our own, only to find out, that the public source code on github did not match the one deployed by the agency. The OpenSource Version on GitHub lagged the version the agency deployed by about a year. Some files critical to deployment were completely missing from the public release. Furthermore the Code was a complete mess. Apparently they never had a successful deployment. The fact that 3 years later in this very thread people are still pointing out that it was used years ago in Taiwan (which was the PoC that convinced the execs at my NGO in 2018) is a testament to that.
If you look at their GitHub (which they don’t promote a lot on their website) they explicitly say:
„ If you'd like to set up your own deployment of Polis, we encourage your to reach out to us for support. We look forward to working together“
This is the real purpose of their project, they try to get consulting gigs for themselves and extract a maximum of money out of unsuspecting NGOs and Govermental Orgs. They have sweet talking sales people but are unable to deliver on their promises. Never Again!
It may be, that the legal issues, surrounding (commercial) YouTube Video production, require getting the explicit consent of the subjects. But this has quite a disturbing effect on the very meaning of what we consider Street Photography to be.
Many "Photography Influencers" that show their process on tiktok or YouTube ask all the subjects of their pictures for explicit consent before taking pictures that I would consider to be more in the realms of fashion or model photography than in the true spirit of street photography. Street Photography is, to me, an art form that requires capturing unstaged moments and not just staged shots outside of the studio. It is the more artisticly inclined cousin of photojournalism, at times prefering form over function and serendipitous beauty over obvious significance and timely message.
BTW It's notable that YouTube in particular seems to steer viewers to channels that celebrate hobbies which if they don't require expensive purchases can at least benefit from them.
The "Photography Influencers" are a niche that owes its growth to the massive advertisement Dollars surrounding the space. The more people watch photography content, the bigger the audience Alphabet can sell to Sony, Canon and thousands of accessoire makers. So there is an incentive to push their almighty recommendation algorithm to steer users into this niche and grow demand. At the same time on the supply side, content producers, are incentivized to go into this niche by the higher availability of niche adjacent sponsorship deals for their videos.
This dynamic obviously plays out in other niches as well, whether it be cars, gardening or golf. Potential sponsors and advertisers for math youtubers like blackpenredpen however are more limited so google is disincentivized to grow their niche. Does this, over the long term, push peoples preferences toward more materialistic pursuits ? I guess we'll find out.
Worthless study. They categorize all time spend in social media as „socializing“ even though tiktok,facebook,and instagram serve up a lot of video content that is consumed passively. Instead of asking participants, what type of media they consumed YouTube vs Instagram, they should have asked what type of activity participants perform on these platforms. Creating tiktok videos is very different from writing YouTube comments or watching Videos on Facebook.
Aggregating all „gaming“ activities together further negates all potential information that could have been gained through this study. Playing Chess online or competing in and training for E-Sports Tournaments will have very different cognitive outcomes from playing candy crush. For a first approximation they should at least have differentiated between casual and non-casual gaming though further granularity would of course be even better (at the cost of requiring larger sample sizes to be meaningful).
Their Crime is not in conducting a worthless survey though. Lack of time and resources might explain that. No, it is their baseless and unscientific „discussion“ of their results that should disqualify this preprint from publication in any serious journal. Speculation is not Science.
While You’re the only one able to judge your personal experience and I can only go by what you have written, I have close Friends and relatives who had similar experiences. Runny nose in particular was not seen as a Covid Symptom in adults by health authorities at the time. Chest Pain is a common symptom of anxiety. All my Friends who had similar symptoms and got antibody tests prior to vaccination had no Covid antibodies. Let me suggest an alternative explanation that might or might not also apply in your case. Constant stress and anxiety about catching a horrific disease or loosing family members lowered the immune response as it is a well known effect of stress. Being fully vaccinated not only lifted that anxiety but also gave the immune system a boost. Thus no more infections with beta-coronaviruses and other viruses that caused the runny nose.
Well if it’s bullshit you should have no trouble providing a counter example. There are tens of thousands of adults learning the rules of chess each and every day. Is there any Grandmaster among them ?
I have worked with adults and children learning chess and regardless of how much time they invested the learning curves would invariably look very different.
Even if you’re willing to devote 10 years of your life to chess you will not become a grandmaster unless you’re starting out as a child. You might be some kind of Ramanujan level chess prodigy, but those are rare these days and in any case You would find out quickly if you had that degree of talent.
The accessibility of chess databases and engines that even the world champions of yore did not have means that anyone with a smartphone can train at the highest level.
The best Ballerinas start dancing quite young, the greatest Chess Players have to start even younger.
The case of the Polgar Sisters [1]is quite instructive. So even if your ambitions to become a GM might be doomed, you’re Children might just realize what you never will.
[1]https://en.m.wikipedia.org/wiki/László_Polgár
Correcting a model is always a somewhat subjective process that can introduce bias, especially when it is done after the study is completed pre-publication.
I am not at all opposed to such adjustments and I was actually involved in a research project that improved on pure chance based randomized selection of trial participants into different buckets (substance/placebo) by weighing in the necessary adjustments in the selection phase. That is why I’m intimately familiar with the state of the art in academia. Novel approaches, such as the one I worked on, are unpopular with reviewers which makes them unpopular with researchers even though they would allow, for faster trials, with smaller cohorts or alternatively to find smaller effect sizes.
My criticism of the OP Study however is not the adjustments per se, which as you point out seem reasonable (although we don’t know what alternative corrections they could have made with the available data and what impact that would have had on their findings)but the fact that they had what was needed to conduct one of the rare properly executed longitudinal studies on this question and wasted that opportunity.
This Study suffers from at least one significant methodological flaw. They did not evaluate the same individuals pre- and post Covid infection but instead evaluated people as a group who were either previously infected or not. While they did try to correct for some socio-economic factors such as age their data did not allow them to factor in the socio-economic discrepancies in the distribution of early Covid cases which overwhelmingly hit service industry workers with direct customer exposure. People with Graduate Degrees were much more frequently able to self isolate and work from home than people without a high school equivalent degree.
What I find to be quite sad, is the fact that these researchers were actually in a position to retest a subset of the original test takers as more than 10% completed a follow up survey in December 2020 which surfaced 275 individuals who had contracted Covid after their first participation in the study.
Retesting those, as well as the >8000 individuals who had no change in Covid status to control for increase in age, increase in points due to getting better at the test with repeated exposure etc, would have been a sufficiently large dataset to actually confirm or rule out at least large effect sizes.
In the best case a cookie banner gives me an opportunity to trust a site owner (not necessarily a company) that they won‘t use cookies to do invasive tracking I didn’t agree to. Oftentimes sites just straight up require consent to „necessary“ cookies in order to use the site at all. In these cases the EU imposed Cookie Banners don’t even give me the option to disable cookie based tracking. In other cases it takes minutes of my time to navigate the cookie banner to get the desired results.
Why should I choose such an ineffective solution if there are superior technical solutions to the same problem such as privacy aware browsers or extensions that don’t require user trust in site owners compliance but stop cookie based tracking on a technical level ?
The Cookie banner legislation actually did nothing at all for privacy, quite the opposite it made browsing the web in incognito mode much more annoying as you’re constantly forced to close cookie banners before beeing able to consume the actual on-site content. (Googles (coincidental?) Decision to disallow chrome extensions in incognito mode renders relevant anti-cookie-banner extensions useless)
You’re correct in that this is a prevailing view in epidemiological/Public Health circles. But Medical Science is not only the macro but also the micro level perspective. Individual Practitioners of medicine might well appreciate more and earlier Data. A first principals based argument is a complementary approach that might uncover things that an empirical view might hide.A good example of this is the Australian Noble Laureate Barry Marshall (https://www.discovermagazine.com/health/the-doctor-who-drank...). As long as the hypothesis can consequently be validated in controlled clinical studies, a theoretical argument even without existing foundation in empirical literature can still make for good science.
Not at all. But since the time this was published many people have been vaccinated and we have a lot of data that the researchers didn’t have at the time of publication. The Speculation in the article, while totally valid at the time, should be replaced by a comprehensive analysis of the available data.
The Problem is not Vimeo, the Problem is that YouTube has established the expectation that Video hosting on the web should be free. No other Video CDN can compete with that outside of a specific niche. Google owns Video.
Sadly most of the research on adapting teaching styles to students learning styles has been completely relegated to special needs education in most countries. The Potential in Neurodiversity is not really appreciated in mainstream Education where it is mainly seen as a disability or hindrance to following a standard curriculum.
But the more we learn about the brain and cognitive development the clearer it becomes, that Neurodiversity is not an aberration but a fundamental feature of human evolution. An educational System rooted in the 19th century based on a fordistic conception of Students and Teachers does not align with contemporary research and will have to be thoroughly disrupted.