Sci-Hub users cost ASA journals thousands of downloads | Hacker News Reader

Sci-Hub users cost ASA journals thousands of downloads

familyinequality.wordpress.com

169 pointsdredmorbius88 comments

For those who support sci-hub's mission and would like to help ensure it can never be taken away, you may be interested in the following project to lay the groundwork necessary for a widely-replicated, decentralized version of the repository:

https://www.reddit.com/r/DataHoarder/comments/ed9byj/library...

All of the documents accessed through sci-hub are archived by the library genesis project and made available as torrents. Currently there are just over 80 million articles included in these torrents. The total size of the archive is around 70TB. The link above also refers to the library genesis books collection, which is 33TB.

This effort has seen tremendous interest in recent weeks; the books collection (libgen) is now widely replicated, but around a third of the scimag articles collection (i.e. those from sci-hub) has only 1-2 reliable seeders and needs more before it can be considered safely backed up. If you have the resources available, I would encourage you to consider assisting. Previous discussion of this project is at https://news.ycombinator.com/item?id=21692222

Looking forward, the next step is to find a suitable way of providing accessible, truly decentralized access, without relying on a single point of failure (i.e. a web interface). Some have been exploring IPFS as a potential mechanism, but there are many ways this could be done. This is a challenging but important problem to solve; with the data available, there is now an opportunity for developers to address the access issues. There may come a day when the sci-hub website goes offline, and it would be good if a fallback is already in place at that point.

Aaron Swartz may no longer be with us, but his spirit lives on. It's now up to us to carry on the fight for which he paid so dearly.

Except everything after "Looking forward," is still an open research problem.

> Some have been exploring IPFS as a potential mechanism

IPFS has no anonymity, so individual nodes could be attacked the way individual torrenters got sued back in the day.

> but there are many ways this could be done

There are currently no scalable ways this could be done. If there were you wouldn't have a single bold researcher providing a single point of failure search engine that hops to some other location each time authorities shut it down.

There is nothing that stops nodes from hosting IPFS content from behind a VPN.

Then you're just moving the point of failure to the VPN.

I wish I could help, but I cannot afford at least 70TB even for myself..

Then this comment isn't for you, but rather other HN readers.

If you believe in the importance of science, and if you're a software engineer at FAAAANG, then you can afford a 75+45 TB (scihub+libgen) hard drive array. If you are making $300,000/year then a $5-10k hobby project to store distilled human progress is something that you could make financially possible for yourself.

Consider doing this, because this might be the last opportunity to get a relatively complete copy. Just having a copy and letting it sit for 10 or 20 years can be hugely valuable to the world, let alone your community.

10TB drives frequently go on sale for ~$160. It may not even cost $10k.

And of course you can partner with some other like-minded folks.

For sourcing drives probably better to go with buying external 10 TB drives (and shucking them). Make a JBOD.. I dunno.

A quick ebay search right now shows used LTO8 drive for $3K (same as new), LTO7 for $1.8K, LTO6 for $0.5K, LTO5 for $0.15K. If you shop around, you can find much better deals.

Here are some tape costs:

LTO-5 (1.5TB/$19.60 = $13.07/TB) LTO-6(2.5TB/$22.58=$9.03/TB) LTO-7 (6TB/$57.95=$9.66/TB) LTO-7 type M (9TB/$57.95=$6.44/TB) LTO-8 (12TB/$134.25=$11.19/TB)

Breakevens:

    LTO8=300T
    LTO7=300T
    LTO6=75T
    LTO5=50T
    LTO4 and below=never

Of course without a tape robot, no one should be using LTO5--there's some personal inconvenience cutoff for everyone.

the data is divided up into torrents of a few GB to some dozens of GB. pick a few that you find numerologically interesting (i download ones with my birth year in their number) and back those up.

It's still extremely raw, but I've done some work on decentralized access to sci-hub. Given an index of scihub ID <> DOI which can be generated from the DB dump made available by libgen https://github.com/frrad/scimag it's possible to selectively download only the parts of the torrent required to get the article you are looking for.

Check it out, contributions welcome. https://github.com/frrad/skyhub

What is the legal situation on this?

I'd love to seed. I even have the space for quite a lot but seeding illegal content in Germany is a big nono.

I think this project is an exercise of civil disobedience.

> but seeding illegal content

the content isn't illegal, the act of copying it might be

An important distinction frequently blurred by copyright maximalists. Thanks for pointing it out.

Aaron Swartz may no longer be with us, but his spirit lives on. It's now up to us to carry on the fight for which he paid so dearly.

Has anyone proposed naming an "information making free" service after him yet?

I like that idea but I'd like to reserve it for something that actually makes impact rather than for someone just starting using Aaron's name. Also, I think you should always contact the family first to ask for permission.

There's the "Aaron Swartz Day" which both Internet Archive and EFF celebrate.

Then there's Star Wars, Swartz always being you you, and such (a joke made with Jonathan Schwartz as well). I quite like that saying, but it may be considered cheap/offending by some.

Those aren't lost sales. Nobody sane would pay $40 or so for a 20 pages paper. Even if Scihub didn't exist. Publishers know nobody pays that. There are other options free for academic, such as subscriptions universities pay for or inter-library loans. Non-academic clients may try asking somebody who has a subscription access for a given paper (I know I did provide paper PDFs a few times when asked).

Publishers don't prey on individual users, they prey on universities. They try to get them to pay for journal access. Non-academic clients? Not profitable enough for us, go away. Scihub threatens this business model, because universities may decide to stop paying that money because Scihub exists. It's much more convenient than bothering a random person to provide you a paper or inter-library loan which makes this an actual risk to them.

The current business model of publishers is predatory and harms science more than it helps, but I see self publishing and private open source initiatives as equally problematic. The definitive long-term solution, in my view, is a centralized public repository, like the Library of Congress. Call it the Federal Scientific Repository or the National Research Papers Database. It would be funded by Congress, or the LoC, and peer review would still occur. But the process of publishing papers in journals with editors and letters to the editor, as well as the environmental and financial expense of printing thousands of paper copies of these publications and shipping them to universities around the country, will cease. The hundreds (if not thousands) of thick volumes on the shelves of university libraries are never consulted by anybody - researchers use the Internet - and take up too much valuable space. It's time the research world catches up.

The adjacent-possible path to that all but certainly involves something that very closely resembles Sci-Hub, to demonstrate that the alternative of continuing proprietary publishing will not function.

Without a Sci-Hub-like option, the political pressure on government (and likely: university) archives to not provide free-to-the-public services will likely continue. Only the continued persistence of a truly free option, in flagrant violation of an unjust and immoral legal environment, can break down that wall.

How on earth can self publishing and open source even be problematic? Let alone equally so as ones whose only role is profiteering by restricting public spread only by if they get their cut or not and providing no validation - peer reviewers.

It's the same "discovery problem" that every long tail market encounters: a significant portion of the tail is going to be either low-quality or irrelevant to most consumers of that domain. So you need aggregators, and those aggregators will face pressure to moderate the long tail by surfacing the most relevant results, while also still providing access to the tail in some form (see also Ben Thompson's aggregation theory [1]).

Unfortunately, this dynamic is more social than anything: there can only be so many aggregators that can occupy the zeitgeist. That scarcity of attention leads to rent-seeking behavior, which is the main problem with academic journals. The current situation is probably approaching optimal; large institutions fund the moderation and aggregation through journal subscriptions, while access is widely available via Sci-Hub.

Without a profit motive, such moderation would be even more politically-driven than it is today. At the same time, the Sci-Hub community does need to ensure its continued existence because it is such a massive boon to science writ large.

[1] https://stratechery.com/2015/aggregation-theory/

The "moderation" part of journals, the editorship and peer review, is already performed virtually entirely by volunteers.

My roommate is a student with full access to basically all journals. Funny enough, he still uses scihub because it is easier to use.

This is a big reason why piracy exists even when something is available through official means.

Those offering a paid services should always ensure there is a benefit and convenience compared to piracy. Valve and Steam wouldn't be here if it was clumsy like in the beginning. Gamers like that getting a game is just a click away, that it installs and checks all the required dependencies, keeps the game file updated, allows you to preload and keep your game saves in the cloud (when possible).

That's a really bad sign for mainstream journals.

At the end of the day when considering the behavior of large numbers of heterogeneous users, UX is king for all data consumption. The pricepoint matters, but people would use SciHub even if traditional journals somehow moved to free publication... If the UX is better for SciHub.

I use sci-hub at home because remoting into my academic network requires me to use 2FA. It's not like the journal is charging my institution per download, so there's no monetary difference between me getting it either way. So I go the lazy route and I don't really feel bad about it.

I also have full access to basically all journals and recently surveyed well over 500 papers in my field of research. Sci-hub was an indispensable tool for this, because it provides a consistent user interface to download _all_ papers. I was able to automate 90% of the PDF downloads because of it, some were blocked because of captcha's, but I was able to solve those manually.

This type of semi-automation would be very impractical and cost an order of magnitude more time if I had to dig through the probably non-existent API documentation individual publishers have, or try and scrape their I'm sure by now heavily obfuscated websites.

I assume most of the users of scihub have legitimate access to some content. Even the most well funded universities don't subscribe to every journal. When doing a lit search, it is very common to wind up with many potentially interesting papers you don't have subscription access to. Getting access to these papers legitimately involves writing request forms for each individual paper (as access typically costs something like $40 per paper) and then sitting around for between a few hours and a few days before being granted access. Frequently the papers don't even contain the data you want (which you can't know in advance because you can only see the abstracts before paying), rendering the entire exercise a waste of time. It's far easier to simply jump on scihub, get the data you need immediately, and sidestep this process.

It's also not like there was no piracy before scihub showed up. When I encountered a paper I didn't have access to back in the day, I would ping all of my friends at various universities / companies / hospitals and request that they check if they had access. If so, I would ask that they download it and send it to me. They would do the same when appropriate. Even doing this was more efficient than going through official channels, and every time this was done it represents another lost sale. From a user perspective, the piracy has nothing to do with money and is instead entirely about friction.

Same. When I heard about Scihub, this was the page I usually downloaded papers from. I used to have access to a lot of journals, but downloading a journal was annoying. Meanwhile, Scihub provided a field to put an URL in, and you got a PDF back - which was ridiculously simple.

Visitors to university libraries reportedly continue to be harassed by campus police officers for "overusing" the taxpayer-funded electronic resources which they are supposed to be making available to members of the general public in exchange for the generous property tax exemptions (under so-called PILOT agreements) that they receive from local and state government.

At MIT, a few years after the arrest of Aaron Swartz, a prominent researcher from Mass General Hospital was threatened one afternoon with arrest by an MIT police officer for the alleged "crime" of using one of the library computer terminals for "longer than an hour."

It's worth remembering then that there are some out there who are using SciHub now because they don't want to go to jail for trying to access the same content through more "legally legit" channels.

>Those aren't lost sales. Nobody sane would pay $40 or so for a 20 pages paper. Even if Scihub didn't exist. Publishers know nobody pays that. There are other options free for academic, such as subscriptions universities pay for or inter-library loans.

This may sound tinfoil hattish, but ever since I was in school i've believed it's to keep knowledge out of the hands of the poor. Accessing journals is basically out of the question for ordinary people, unless they go to a university library, or spend money they're not going to spend, almost all reporting on science is complete and utter garbage that at best, gets most of the facts wrong, at worst misrepresents or distorts the article being reported on completely.

I don't like making blanket statements, but i've noticed, many people lack even basic knowledge of scientific methodology and almost see it as a black voodoo magic or something. Average people have no real way to verify scientific claims made by media even if they wanted to.

I have to be honest too, going through school, this attitude was prevalent among my teachers, that scientific knowlege was best left to scientists to interpret for average people. Everything around science has a 'it's our special club' thing going on designed to trickle only what information scientists and authorities deem necessary.

Hell, the data from my own project was locked away by the government behind a paywall, despite it being funded by tax money under the stipulation in our grant that our results and data would all be made publically accessible and we were only studying small mammals.

Science, especially publically funded science, should be available for everyone. I've never agreed with the idea of hidden knowledge. Human progress comes from humans learning and sharing knowledge with eachother. Keeping knowledge locked up for a special elite cripples the progress of humanity.

poor != unintelligent

That said, I get the concern that non-scientists can often misrepresent scientific findings. I read the other day that apparently lemon peels are 10,000 times more effective than chemotherapy at treating cancer for example. But I don't think the solution is to lock away knowledge. The solution should be to try to educate people as much as possible.

>But I don't think the solution is to lock away knowledge. The solution should be to try to educate people as much as possible.

Yeah...me too...I'm guessing people didnt actually read what I wrote...

I did read what you wrote but your position was somewhat unclear. Your first several paragraphs seemed to conflict with your final statement.

Unfortunately, providing access to everyone would be incompatible with publishers' business models. As long it's more profitable to not provide the access the everyone, business will continue to do so.

Journals cost the general public access to science while having an even more questionable value than most copyright holders who at least bother to do things like remaster their old properties or develop new drugs. May they burn in hell.

Sci-hub means I can read the latest science about my disability without $40 paywalls on top of a mediocre paper which give eye-poppingly low margins to the actual authors. Papers which ironically often discuss how society makes disability a problem by throwing unnessecary barriers in front of the disabled. That yarn on about how the disabled lack autonomy over their own healthcare.

Zero margins. Scientific authors get exactly zero money from academic publishers. In fact, sometimes it's negative money (the journal charges you to publish, i.e. if it's open access). Zero or negative. They get exactly zero dollars of that $40-per-article fee. Which is why authors will often GLADLY give you the article for free if you email them.

Scihub is in a legal grey area (not really legal, so I can't technically advocate for its use in my position), but in a way it's automating the normal process of asking researchers for a copy of the paper, thus saving the researchers' time. (Researchgate is a more legal version of this, but with much less access to papers.)

> In fact, sometimes it's negative money (the journal charges you to publish, i.e. if it's open access).

Not just if it's open access: sometimes authors have to pay "page charges" or have to pay extra to include illustrations.

> which give eye-poppingly low margins to the actual authors

You mean exactly 0? Authors are almost never paid to publish in a journal.

,,I do not, however, contribute anything to the system; I free-ride off their criminality''

The writer has to constatly remind me that running the proxy is a crime, but using it - even if she knows that it's not the legitimate way to get the information- is not.

This may be true, still I don't like the tone, as clearly sci-hub is providing the huge value to people, not her.

I wouldn't be surprised if some life saving drugs would be developed and used in practice faster because of sci-hub.

I don't use sci-hub, but the framing "Sci-Hub users cost ASA journals thousands of downloads" presumes something: it presumes that without sci-hub, ASA journals would have thousands of more downloads.

This could very well be false: without sci-hub, people might instead use arxiv and the like; perhaps sci-hub brings journals thousands of downloads they would not have if sci-hub didn't exist.

As I don't use it I can't say if this might be accurate, though.

Or, people might not have published to ASA journals if their papers would not have been easily accessible from sci-hub.

I use it a fair bit for a variety of purposes. Without Sci-Hub, given the astonishingly high pricing usually involved I certainly wouldn't be paying for the articles. I'd be managing without as best I can.

Sci-Hub doesn't give me information for free I'd otherwise have to pay for - rather it provides information I'd otherwise have little access to.

We went through this with MP3, record industry claimed every download was costing them money, that is was the end of the world as we know it, they would wither and die without copyright protection, while at the same time, more people were listening to more music, and look where we are now...

Didn't iTunes store used to be DRM free? (I never actually had any apple device so I'm not sure)

Other way around - it used to have DRM and removed it a few years in

>This could very well be false: without sci-hub, people might instead use arxiv and the like; perhaps sci-hub brings journals thousands of downloads they would not have if sci-hub didn't exist.

There was that study in EU that said piracy actually helped the game industry.

There is also what the author was commenting, sci-hub is easier to use than the legal alternative. We go tour researchers stuck in bureaucracies, to access papers funded by taxpayers! As Gabe Newel said, to fight piracy, you provide a better service (Steam was the biggest hit to game piracy, and so was Netflix for series and movies), instead of that, what we see is more control/bureaucracy being added.

edit: https://www.engadget.com/2017/09/22/eu-suppressed-study-pira... it's an old study (2013), but I doubt the situation got worse. In videogame land the greatest example is The Witcher 3, easiest game to pirate, has it was distributed without DRM, you could just copy your friends installer. It was still on the top of steam most played single player game(which requires the key to be unique), won a lot of awards, and got so much exposure that netflix has released a TV show...

HN is getting novelty usernames now? That was a bit, um, unsettling.

Well, I didn't want to associate pro-piracy views with my main account. That account should be for tech discussion only, without politics. Identity on the internet is forever.

It's hard to imagine how a legal service could be better than Sci-hub. You could make it as good as sci-hub, but how could you make it easier than "paste DOI -> PDF is downloaded"?

Steam beats video game piracy because video game piracy requires shit like having a torrent client or having a friend who has a copy already. Video game piracy was never as easy as sci-hub.

Just make it _as good_. Add legitimacy at the same level of convenience and you win.

I'd pay $10 a month if addition to pasting a doi -> pdf, that I could also search backward and forward along citations. I'd love to be able to download a paper and everything it references.

Iirc that's pretty much why TBL invented the web, so you could click on references in scientific papers. It'd be nice if we could actually do that.

I'd probably be willing to pay $30/month for that. Of course they would have to actually have all or most of those backward and forward references available.

As other commenters have responded, being able to navigate the DAG of citations would be a nice value add.

I think that the existing journals will always lose this unlikely publishers. The publishers at least contribute to the content which people appreciate. The journals are only parasitic. Plus we shouldn't pay the paper publishers, although harmless for some fields in others that would make the peverse incentives /even worse/ as it encourages what the audience wants to hear.

Of course you're right, but the article acnowledges that at some point.

I'm not a researcher, but download lots of articles just as general curiousity. I definitely wouldn't have $30 for each download, as I'm not profiting of it at all.

Requesting a song from a DJ so you can tape it from the radio for free costs record labels millions!

To be frank when people using "..., and that’s OK" they pretty much lost me. It implies the person thinks their value judgement on the matter is somehow universal - which it just is not.

I use sci-hub - others can decide for themselves to use it - but I'm not going to say others should think it is okay or not okay if I use it - the value judgements of most other people regarding my actions is not something that I care about a lot.

> I wouldn't be surprised if some life saving drugs would be developed and used in practice faster because of sci-hub.

Seems like it'd be difficult to precisely estimate the effect of open literature on advancing education, research, and development efforts.

Back when I was a student, we had access to a lot of journals -- but not all of them. And to use them, we had to connect through a laggy VPN to the university, as we often traveled; sometimes it was down. And then when we worked with industrial partners, we couldn't just email them relevant papers as that'd have been illegal. And we couldn't even just email each other papers, as I think the terms-of-use required us to independently download articles.

And, again, that was with university subscriptions! Rising students and professionals who go into industry typically can't access articles at all. Sure, if they really need a specific paper they can try to find a way, but that's a horribly inefficient way to do research; to be effective, a researcher really needs to be able to browse through the literature freely, reading lots of papers without being distracted by technical barriers.

Given that the current system slows down everyone, it's hurting all of us. All of science, technology, engineering, medicine, whatever -- it's all less than what it could've been because we're being held back.

But how do we quantify the on-going harm? How can we guess what sorts of inventions weren't made because researchers didn't have access or were fighting with a buggy VPN? How do we guess at how much education's being held back by students not knowing how to access the literature before college? What about all the damage from working professionals not being able to browse the literature related to their field?

Anyway, yeah, the article's constant references to Sci-Hub, as though it's somehow criminal, were definitely jarring. Sci-Hub feels like it should be legal based on the same sorts of arguments that make self-defense or providing emergency assistance legal.

Sci-Hub helped me identify the appropriate life-saving drug for my disabling illness. I could barely sit up and now I'm walking around. It's going through FDA approval very slowly, so I just buy it from a manufacturer from China.

My university gives me access to many journals but i still use Sci-Hub. I will never download adobe digital editions just to view a pdf.

And last I checked it does not work on wine and there is no Linux version.

It does work on wine but requires some work to get it running: https://appdb.winehq.org/objectManager.php?sClass=applicatio...

Using Sci-Hub is still faster and easier than logging into the university VPN or web proxy (which breaks HTTPS sites).

might be worth looking in to split tunneling if you can do it. my other solution for this at a previous job with an annoying multi-network setup was to have a vm run the vpn and then squid proxy anything i need out to the main host.

Even though I have access to the full Stanford library collection I go to Sci-Hub first. The UI is so much easier.

Is there seriously anyone not employed by the giants (Elsevier, Bertelsmann et al) who opposes sci-hub?

Maybe I would care about the plight of academic publishers if I got royalty checks for all the articles I've published over the past 20 years.

Be careful with piracy in the US:

It’s not enforced by police but if you’re doing science or engineering then the biggest employer in your field might be the government (DoD etc.) It’s not all that rare for people to get denied security clearances because of excessive piracy and that would make it way harder for you to be employed.

This is a huge part of the problem with having laws on the books that just aren’t always enforced.

For context, the author is an American sociology professor who serves on the American Sociological Association's committee on publications.

He recently came out in opposition to the ASA's official stance rejecting a proposed federal mandate to make all publicly funded research available open access. The ASA joined other scholarly associations in opposition such a mandate.

The author recently discussed his stance on The Annex podcast: http://sociocast.org/podcast/the-paywall-and-the-asa/

Good, fuck the journal aggregators. Speaking as someone who has published papers, I'd happily (and do, via ResearchGate) give them away for free. I made the noob mistake of putting them up on my own website before being shouted at by an editor for giving away the 'product' for free.

I'm all for journals, I appreciate there's a cost to keep them going but large publishing aggregators who take your work for free and have the cheek to make you jump through hoops for the privilege deserve to collapse. The paradigm is old and broken.

Oh no, who will think of the poor journal editors' profit margins?

Sarcasm aside, that title is waaaay loaded. It sort of insinuates sci-hub users are the burden and not the journals.

Times change. Sometimes laws -even once good laws- can end up becoming obviously absurd. When that happens in a democracy, the laws can be changed.

With all the people using Sci-Hub, I wonder if it is becoming possible to make legislation to either or both

* Declare Sci-Hub and its practices legal outright, and make it the new normal.

* Carve out extra protections for open access to science in general.

Also give the woman in Kazakhstan behind it the peace Nobel price.

> * Declare Sci-Hub and its practices legal outright, and make it the new normal.

It's not exactly what you mention, but most Dutch researchers are legally allowed to freely share their work six months after publication: https://www.openaccess.nl/en/in-the-netherlands/you-share-we...

> Article 25fa of the Copyright Act (Taverne Amendment) allows researchers to share short scientific works (e.g. articles & book chapters), regardless of any restrictive publishers' guidelines.

> The maker of a short scientific work, the research for which has been paid for in whole or in part by Dutch public funds, shall be entitled to make that work available to the public for no consideration following a reasonable period of time after the work was first published, provided that clear reference is made to the source of the first publication of the work.

Oh god! What if we made sci-hub legal though? We might usher in a golden age of science at the cost of short-changing parasitic middlemen!

Consider also the value of SciHub that you have received already: It has made the authoring of Review papers much easier, and probably improved their quality since you can parse hundreds of papers faster. No longer do I have to worry if the publisher's site is accessible from home, if their website is not down, if i'm gonna have to click 10 links to get the pdf etc etc. A click of a bookmark and PDF is downloaded. Literature reviews are already hard as it is (with articles being pdf and unhyperlinkable), paywalls just made it double as hard.

> (with articles being pdf and unhyperlinkable)

I'm not sure to what extent this helps, but it's at least possible to hyperlink to an individual page in a PDF.

https://isotropic.org/papers/chicken.pdf#page=2

unfortunately, bibliography , which is the most important kind of link is almost never hyperlinked.

An increasing number of papers in my field have the DOI the cited paper as a hyperlink to dx.doi.org which will redirect to the publisher website. Note that this is often done without optical hints (text in black not blue, no underline) so it can be hard to miss.

Maybe these journals should get together and duplicate the Scihub interface which pulls from whatever sources the user should have access to.

I have never used these services at a University. Do they even require a login? Or do you get automatic access just from using a terminal on campus?

Normally you get access automatically when you're on any machine connected to the campus network. Most schools also have a proxy server or VPN you can use off-campus. The Zotero browser plugin does a good job for me of automatically detecting resources that can be proxied and redirected so things work pretty seamless for me off-campus as well.

Thousands? Millions more likely!

Journals and papers should be obsolete. New results should be entered into an AI system containing the existing body of knowledge on the subject and which can rapidly evaluate whether the new result is consistent with, contradicts or extends existing knowledge. It would also evaluate plausibility, reproduceability, methods and materials, plagiarism, etc.

Haha if that existed the AI could just emulate infinite monkeys on typewriters to create papers it could then filter for the plausible, reproducible ones to publish. Hm, maybe it would have to adversarially train the emulated monkeys into monkey researchers to get better throughput.

Seriously though our beloved AI doesn't have reliable object permanence yet. It can't read now and it certainly won't be reading research papers for a while. Be nice, give it some time to grow.

I for one welcome our new AI overlords!

Reply on news.ycombinator.com