Science, culture, complexity

Tag: Digital Millennium Copyright Act

  • A new way to harass editors?

    There’s a new way to harass editors – or perhaps it’s an old way that we’re just finding out about, first-hand. We know that repressive governments have started using the US’s infamous Digital Millennium Copyright Act (DMCA) as a new means to censor content they do not like. But it now seems private parties have also discovered the utility of using alleged copyright infringement to target media coverage.

    In the first week of October, an individual with an address in Thane, Maharashtra, lodged a complaint with Amazon Web Services (AWS), which hosts The Wire websites, including The Wire Science, alleging that one of the latter’s articles contained ‘unlicensed copyright protected content’.

    It has been my experience, and that of every other editor, I imagine, that honest complaints of copyright infringment are addressed to the editor and the reporter in question – and not the website’s host. (My email ID is on The Wire Science homepage; our ombudsperson’s email ID is available on the ‘About’ page.) But in this case, the complaint was lodged with AWS, with a link to the corresponding article on The Wire Science.

    The AWS abuse team, in turn, has written to me and my colleagues multiple times asking us to specify what steps we have taken or will be taking to resolve the issue. We have written back but I am not sure if the members of the abuse team are equipped to understand the editorial issues involved. Their principal issue appears to be that the charge implies The Wire Science has violated AWS’s terms of service and could therefore have to be removed from its servers.

    Why would the complainant take this route to resolving an allegation of copyright infringement? The article in question could provide the answer: it is an investigative report by science writer Anusha Krishnan (April 3, 2021) about a device called ‘Shycocan’, whose makers have claimed it can “attenuate” particles of the novel coronavirus by simply emitting photons into the air of a room. The report cast doubt on ‘Shycocan’ as well as its maker’s claims.

    The news report

    One particularly important, but easily refuted, claim made by the company was that the device has the approval of the US Food and Drug Administration (FDA). The truth comes in two parts: that the FDA found the device to fall in a category that could be distributed in the US without complying with certain regulatory requirements, and that according to ‘Shycocan’’s maker, the device belongs in the category of ‘sterilisers and disinfectants’, not ‘medical devices’ per se, meaning it doesn’t need clinical trials to prove its merit. Some claims had simply spun these loopholes in the maker’s favour.

    The maker’s representatives responded to our article on ‘Shycocan’ with a detailed statement sent to me seven weeks after the article was published, even though both the reporter and I had asked them many of the same questions during reporting (with many days to reply). I refused to publish it because I had no obligation to do so – plus it seemed to me to contain unclear science.

    For example, Umesh Kadhane, the head of the physics department at the Indian Institute of Space Science and Technology, Thiruvananthapuram, had told The Wire Science that the maker’s “claims that the electrons produced by their device will only kill the coronavirus is completely bogus. Electrons cannot distinguish between viruses or bacteria or any other thing.”

    The statement responded thus: “Most bacteria are negatively charged. The Coronavirus is a positive sense virus. Human cells have a negative potential. The electrons emanated by the Shycocan and due to this opposite polarity they attach themselves to the Coronavirus in real time thereby neutralizing or attenuating it.”

    By this logic, ‘most bacteria’ should not be able to affect human cells, so this part of the statement is likely incomplete. The electrons should also be affecting every susceptible particle they encounter once they are emitted – not just the coronaviruses in the air. Many atoms and molecules in air capture free electrons. As the report also said:

    “It’s … unclear how ‘Shycocan’, though capable of producing so many electrons – much more than air ionisers that are currently in the market as air purifiers – apparently doesn’t produce ozone, according to the company marketing it. When oxygen in the air encounters free electrons, it becomes ozone.”

    But the maker’s statement said our sentences lack “scientific basis” and that the device produces photons in the “trillions per second, which in turn produce photoelectrons when striking solid surfaces.” This raised two further issues:

    1. Photons need to have a specific energy to produce photoelectrons from specific materials, called photoemissive materials. Not all photons can elicit photoelectrons from all surfaces (“aerosols, microscopic impurities, viral particles, solid surfaces, walls, etc.,” as the statement says). 2. If all these surfaces are photoemissive (unlikely), why don’t their photoelectrons lead to the formation of ozone? And so forth.

    The Wire Science report also raised concerns about missing details in the documents ‘Shycocan’’s maker shared with us (requesting us to not share them publicly; we agreed). Independent experts Anusha Krishnan and The Wire Science spoke to said the documents lacked information about the testing methods and, in at least one case, efforts to eliminate bias.

    Arindam Ghosh, a physicist at the Indian Institute of Science, Bengaluru, had also told Indian Express last year: “The claim is that if you fire a lot of electrons, it affects and deactivates the S protein of the coronavirus. I do not know of any scientific document which proves this. To my understanding, there is no data available even outside the publication domain which proves it. I do not understand … how some weird electrons seek only the coronavirus to kill, leaving everything else unaffected.”

    As a result of these findings, The Wire Science report said: “As of now, there appear to be no published scientific studies, experiments or publicly available data (that other scientists can use) to establish the efficacy, safety or usability of ‘Shycocan’. All of the information on the device is to be found on the company’s website, in news articles, press releases and anecdotes related by the people marketing it.”

    The implicated para

    Now, the complaint with AWS alleges that The Wire‘s report contained one paragraph in the article that was copied from a magazine with a single ‘article’ hosted on Issuu. The article, which carries the date of March 30, 2021, is published by ‘jaiprakash36’ (who has published nothing else) and  contains precisely two paragraphs – one, the supposedly copied para, and two, a para that appears to be an advertisement for ‘Shycocan’:

    Shycocan Stands for Scalene Hypercharge Corona Canon you can buy it on amazon at the price of 24,999. According to the device makers Shyconcan can disable upto 99.9% virus in the installed area.

    Turn now to the paragraph which the complainant claims was copied:

    The company Eureka Forbes has also advertised the “Forbes Corona Guard, powered by Shycocan” as a device that could attenuate 99.9% of coronaviruses in enclosed spaces. In November 2020, after complaints from scientists, the Consumer Complaints Council of the Advertising Standard Council of India directed Eureka Forbes to withdraw its claims. Yet the company still lists the product as available, along with its purported effectiveness against the novel coronavirus.

    Shortly after receiving the first communiqué from the AWS abuse team, we responded with a timestamped document that clearly shows Anusha Krishnan and me co-editing a Google Doc document containing her report, with edits of the concerned paragraph dated before the Issuu ‘article’ was published, on March 30, 2021. (I suspect, sans proof, that the complainant published the single-page magazine after our article was published and then backdated the page).

    However, the AWS abuse team has been repeatedly emailing us asking us to describe the steps we will or are going to take to resolve this issue, unmindful of the proof we have provided. To them, it appears, this is a potential DMCA violation that can only be resolved by us responding to the complaint by making some changes at our end. To us, the abuse team doesn’t seem to be prepared to consider that the complaint is baseless.

    This has been a frustrating experience, and is still yet to be resolved.

    The Wire
    October 16, 2021

  • Will this blog be online a hundred years from today?

    For almost two weeks now, we at The Wire have been dealing with a complaint that someone from Maharashtra lodged against us with Amazon Web Services (AWS), our sites’ host, for allegedly copying one paragraph in one article sans consent from a source that the complainant allegedly owns, and thus violating AWS’s terms of use and becoming eligible – if found guilty – to have the offending webpage taken down. The paragraph is not plagiarised (I edited and published it) but the alleged source of the ‘original’ material is shady, and there’s reason to believe a deeper malice could be at work, as I’ve explained in an article for The Wire.

    The matter is still unresolved: the AWS abuse team has been emailing us almost every day asking us to tell them what we’ve done to ‘address’ the complaint, ignoring the proof we sent them showing that the article couldn’t possibly have been plagiarised (we shared the Google Doc on which the article was composed and edited from scratch, with date and timestamps). The abuse team remains unsatisfied and would simply like us to act, whatever that means. From my point of view, it seems like AWS doesn’t have the room to consider that the complaint could be baseless. I also think that any organisation that doesn’t know or want to deal with editorial complaints shouldn’t receive editorial complaints in the first place. Otherwise, you have a situation in which an unknown private entity can allege to a tech company that one of its clients has violated the Digital Millennium Copyright Act (DMCA) – the overreaching American legal instrument that is the principal blunt weapon in this episode – forcing the tech company to bear down on the client to make the problem go away, without pausing for a moment to think if it’s been conned into becoming an agent of harassment. But why would it, considering so many tech companies registered in the US actually benefit from the overreaching character of the DMCA.

    (It’s doubly ridiculous when these agents are Indian and based in India, and who may not be aware of the full story of the DMCA but are required by their contracts of employment to enforce it.)

    The awfulness of this entire episode, still ongoing, strikes to me at the heart of questions about who gets to access the internet and how. The AWS abuse team has told us on more than one occasion that if the matter isn’t resolved to AWS’s satisfaction, they will have to remove the offending webpage from their servers. Obviously we can find a new host, but whichever host we find, the problem remains: one of the many mediators of our access to the internet, starting from the internet service provider, is the entity hosting those websites. And not just any entity but a predominantly American one, and therefore obligated to enforce the terms of the DMCA. I don’t have to point to any numbers to claim, safely, that a vast majority of the internet traffic today pings the servers of websites hosted by AWS, Google Cloud Platform and Microsoft Azure. Recently, Tim Bray blogged about what the consequences might look like if the biggest of AWS’s 24 datacentre “regions” – called simply us-east-1 – went offline. It would be an unmitigated catastrophe.

    My blog has been hosted with/on WordPress for 13 years now and I’ve seen a lot of competing platforms come and go.[1] (My very first blog was hosted by Xanga before I moved, a few months later, to WordPress.) A lot of people who like to talk or blog about blogging have expressed dissatisfaction with how some platforms “don’t talk to each other”, that they’re fans of the Quiet Web[2] or that static sites are the way to go for the speed, security and controllability. But to me, all these concerns pale compared to the question of whether a platform will actually stay online. One alternative I’ve been referred to is micro.blog – looks nice and has an agenda that some bloggers seem to love, but will it stay online? I don’t know. It’s easier for me to believe a) that WordPress will stay online because it has been online for 15 years now – which is a long time in the Internet Universe – and because it has been both profitable and conscientious about what it does; and b) that AWS will stay online because its market capitalisation and revenue mean it’s just too big to fail at this point (as Bray has also written). Heck, of all the blogging platforms that have come and gone, one of the longest-lived has been Google’s Blogger. Google clearly didn’t spend much time on it after a point but Blogger is still around, as are the bloggers who continue to publish there. And to my mind the ‘Persistent Web’ – a place that convinces you that it’s going to be around for a long time – is a better place to be (see ref. 2).

    [1] One of the platforms that I was really sad to see die was Posterous, which Twitter bought from the guys who built it and then killed it. These guys subsequently created Posthaven, and pledged that it would never get bought or be killed as long as at least one blogger paid to use it – except the guys have added no new features since 2017 nor updated the blog. Tumblr is pretty much a ghost city now even though its been bought by the foundation that runs WordPress, and even though it attracted a great deal of negative attention for its erotic blogs. Typed, made by a company that became profitable by building apps for Apple devices, was ridiculously short-lived. Silvrback‘s founder sold it a few years ago to some mid-level management professor who’s modified it – badly – back to the early 2000s. Svbtle is still around, and has a pledge like Posthaven’s, but is both extremely minimal in terms of its features and fairly opaque about how it’s doing as a company; also, it appears to be hosted on AWS. Medium‘s mood is just awful and doesn’t seem trustworthy as a company either, especially if you’re particular – as I am – but about not putting in with enterprises that treat editorial people badly. This is also why I’m put off of Ghost, whose maker John O’Nolan has seemed quite full of himself on occasion. Ghost itself is a great product, although it started off as a blogging company only to change direction to become a publishing company, leaving WordPress – which it sought to usurp as every blogger’s platform of choice – to dominate the blogging space. Squarespace is the sole long-lived, equally legitimate alternative to WordPress, but it doesn’t offer a self-hosted version. I could go on.

    [2] Brian Koberlein writes here that he defines the Quiet Web thus: “Exclude any page that has ads. Remove them if they use Google Analytics or Google Fonts. Remove them if they use scripts or trackers. It’s a hard filter that blocks the most popular sites. Forget YouTube, Facebook, Reddit, or Twitter. Forget the major news sites. So what remains? … Most personal websites don’t pass the test. They are either ad-driven or managed on platforms like Blogger or WordPress. But the quiet personal sites are diverse and interesting.” I think he overlooks a huge part of the internet here, comprising websites that self-host WordPress instead of using it on WordPress.com. Many of the features he ascribes to the Quiet Web can be built with WordPress – including this site you’re reading. As celebrated web-designer Jeffrey Zeldman tech blogger John Gruber has written, WordPress shouldn’t be blamed for many of its users’ awful design choices. (Edited at 8:43 am on October 27, 2021, to attribute that comment to Gruber instead of Zeldman.)

    So when someone asks if my blog will be online and accessible a hundred years from now, I’d like the answer to be ‘yes’. And while I don’t like it, AWS is going to remain one of the options to make that happen with little effort on my part. I’m not a programmer except in the tinkering sense, and still struggle to understand how websites really work (esp. beyond the application layer). Pertinently this means I’d much rather host my blog, which is invaluable to me, with an entity that knows what it’s doing rather than try to cobble something together myself that could well break or be exploited a day later. And this in turn is why I’m really going to stick with WordPress, which continues to be an excellent alternative to every other similar option, with my fingers firmly crossed that the people managing it continue to do so the way they’re doing it currently.