rspeer

> Are you talking about someone crediting a webpage without attempting to attach the name of the author? I have never seen this in academic research, ever. Can you link to an example?

Let's continue with the Google Ngrams example. All I had to do was search Google Scholar for "Google Ngrams" to find easy examples on the first page of results, like this one: http://www.aclweb.org/anthology/W14-1611

The paper is written by famous Stanford researchers, and they do not cite a single author of their data. Just URLs in footnotes.

This also happens to my research (ConceptNet) all the time. The fact that you haven't noticed it doesn't mean I'm making it up.

I don't see why you think researchers not giving credit for data is a different topic from what I brought up. What I said was, researchers in my field are in the habit of not giving credit for data, and instead they have adopted the convention of using half-assed footnotes with URLs in them, which is not appropriate credit. I speculate that one reason for this habit is that data often comes from inconveniently non-academic sources and would look bad in the bibliography. The habit is so pervasive that authors remain in the habit even when a paper exists.

Crediting a web page in a footnote is almost no credit at all. The web page didn't compile the data. A person did. Sometimes the person is inconveniently pseudonymous, but other times they should be credited by name.

I added the Google Books Ngrams example slightly later, but: People _do_ publish papers about their data, and their citation counts are often extremely low, because research users continue to use footnotes instead.

It's not that nobody knows how to cite it, it's that nobody wants to cite things from outside of academia, perhaps because it lowers their own perceived prestige.

I have seen this in computational linguistics where people frequently don't cite their data sources, which often come from outside of the field or outside academia altogether. Instead they put a link to a webpage in a footnote, which does nothing for assigning credit.

Reviewers don't take anyone to task for it, because they do that too.

For a while, there was a collection of multilingual word counts over OpenSubtitles that everyone was using, and you would acquire it from the OneDrive of the pseudonymous blogger "Hermit Dave". Many people put the link to his blog in a footnote, but I think I was the only person to come up with a citation for (Dave 2011).

People don't even cite Google Books Ngrams and that's got a real paper [1].

[1] Lin, Y., Michel, J.-B., Aiden, E. L., Orwant, J., Brockman, W., and Petrov, S. (2012). Syntactic annotations for the Google Books Ngram Corpus. Proceedings of the ACL 2012 system demonstrations, 169-174.

I read that page and looked for the reason I don't use Conda (because I already have virtualenvs and I'm not prepared to burn them all down):

> Myth #5: conda doesn't work with virtualenv, so it's useless for my workflow

> Reality: You actually can install (some) conda packages within a virtualenv, but better is to use Conda's own environment manager: it is fully-compatible with pip and has several advantages over virtualenv.

> [...] the result seems to be fairly brittle – for example, trying to conda update python within the virtualenv fails in a very ungraceful and unrecoverable manner, seemingly related to the symlinks that underly virtualenv's architecture.

Doesn't sound like much of a myth then, if Conda's take on virtualenv is "you can technically do this, but everything will break ungracefully and unrecoverably, so please don't".

What would requirements.txt even _do_ in a library? AFAIK that file isn't even read when packaging, distributing, and installing libraries.

Those industries have stringent regulations on them.

When an aircraft has a bug, there is a federal investigation.

The headline is sensationalized -- I understand there's more to the show "American Ninja Warrior" than jogging up three large steps and not tripping over a log -- but the demo video is still pretty cool.

Does that provide the amazing experience AirPod owners talk about?

Because it sounds like it provides, y'know, Bluetooth audio, which is usually a terrible UX.

Are Airpods going to work with the Pixel 3?

Or will it be the more typical thing with Android accessories, where you have to buy a Google knockoff and never lose it, and then when you do lose it, you have to buy a seriously off-brand thing from China because Google has lost interest in products they sold a year ago?

I cannot imagine being so enmeshed in the world of targeted advertising that I would care about this analogy.

I believe you are responding to the tongue-in-cheek title, and not to the article (which is about a security hole in Google Meet).

For the sake of the thread, you could read the article or have some awareness of the topic instead of assuming.

> “The ads themselves are illegal,” Galen Sherwin, an ACLU lawyer, said. “It’s been established for five decades.”

Programming was a women's job, almost exclusively, until the last half century when it became a hobby and a lucrative profession. Men pushed the women out on both fronts.

You are no Grace Hopper, and you are in no position to tell women what they can't do.

Thank you for saying this.

Unfortunately, now you know what HN thinks about gay and trans people having equal rights.

This has somehow turned into your opinion on energy use (yes, it's good that desktop PCs can now decrease their energy use significantly while idle, which was not the case in the 90s).

But you made a claim that "this was never the case", when clearly there was an era of home desktop computing when your desktop could act as a hobbyist server. This era gave rise to BBSes, then MUDs, then Minecraft servers (and many things in between).

Of course it was the case. What do you think things like Seti@Home were running on? People who owned computers had so many idle processor cycles that they were desperate to find something to do with them.

If 24/7 home computers were never a thing, what did BBSes run on?

But the key difference remains: there aren't many personal computers that we leave plugged in and turned on 24/7 anymore.

> copyright on interpretation is about the same category as copyright on a work of a plumber in my apartment

I'm going to guess you're not particularly into Bach, if you're going to make a statement devaluing classical music like this. The interpretation of the performer has always been hugely important. Nobody gets into Bach's music and then listens to MIDIs of it.

It's not like just anyone can put a video on AllOfBach. That site is entirely the Netherlands Bach Society's project.

If you're saying everyone should make their own AllOfBach... the website clearly has taken tons of development effort, its first iteration was unusably difficult to navigate, it's still not great at accessibility, and if I wanted a portable version of their recordings I have even fewer options than I would with YouTube.

I'm glad AllOfBach exists but there's a reason people want to stick to the music services they are already familiar with.

Pointing out the uncivility in your reply is exactly what dang just did.

You should not expect a sentence-by-sentence accounting of where you lost civility points. That's not how it works.

In the comment you're replying to, I said that the quality of any particular spam filter used to matter, and doesn't anymore.

In 2006, Google definitely had the best spam filter. Now, instead, it has the harshest consequences to being caught spamming.

I check my spam folder out of curiosity once in a while, and there is nothing in there I would miss. Like some of it is "legitimate" but it's still just bulk drivel, nothing that obligates me in particular to read it.

Right now, I see mail from Google in my filter. Because it's automated dumb shit where Google Photos wants me to look at my photos from a year ago, and I already told Gmail that was spam, and it believed me.

If I miss an e-mail because it was caught in my spam filter, that's the sender's problem, as opposed to pre-Gmail, when it was my problem.

It seems to me that I haven't _had_ to re-train my spam filter by rescuing important ham messages from the spam folder in a decade. Perhaps your experience is different.

This sounds like satire, but I don't get the joke.

> Email spam has not been a problem since we figured out Bayesian filtering.

My understanding is, this is not because Bayesian filtering solved the spam problem forever. It's because Bayesian filtering solved the spam problem for long enough for Gmail to dominate email.

Google had a ton of data and they were already using it to recommend ads, so they were well-positioned to use it to make the best spam filter. Unlike the things that came before, you could rely on Gmail's filtering. You didn't have to sort your spam folder ever. This was a clear advantage bringing even more people to Gmail.

Now the thing discouraging email spammers isn't any given filter, it's the fact that Google owns email and the cost of messing with Google is too high.

(I am not saying that this, in particular, makes Google evil. Another way to look at it is, Google saved email as a usable medium, by centralizing it.)

Okay, but don't get the illusion that things like Mastodon currently let you port your identity and your data.

Moving from one Mastodon instance to another involves starting over from scratch, just as much as moving from Twitter to Mastodon does. Federated portability sounds like a nice goal and I hear that some people are working on it, but nobody really knows what form it will take.

I mean Jupyter doesn't really do that either, right? You can't edit most function definitions. Only ones that you've defined right there in the notebook, and only because you can run the cell that defines them again, overwriting the function.

But if you do use libraries instead of just a huge mess of notebooks, you're stuck if you want to change the code in a well-supported way. You can ask your notebook to monkey-patch the code (which is an even bigger mess), or you can use an unreliable magic extension, "%autoreload".

Even if you don't personally see the benefit of a platform having a way to stop hate speech and harassment (we can all agree on the value of stopping spam at least), you still have the problem that a platform that allows hate speech and harassment will self-select for horrible users who aren't allowed on the major platforms. It'll be a wasteland of garbage, like 8chan, Gab or Voat.

Notice that you're currently expressing yourself on the actively-moderated HN, not on one of those.

Indeed, I tried to learn how to make a Debian package. A proper one that followed the guidelines (even though I wasn't trying to get it into a repo), not just one that's shoddily converted from another format.

The tools are just a bunch of hacks designed for the particular environment and habits of the few people who make Debian packages. They don't even try to document how an outsider would learn to do it.

Not sure why you added systemd to this, though, except to appeal to the old guard of sysadmins who don't like systemd and show up on HN a lot? It seems a completely unrelated conversation.

Many of these homeowners, unlike "sovereign citizens", actually appear to have the law on their side. They aren't just making up their own insane interpretation of the legal system. It even sounds like it's more like MERS who made up an interpretation of title that disagrees with state laws.

You seem to be saying that MERS's interpretation should be correct, that MERS's database should be the law of the land and investors should be able to buy and sell the title to mortgages without states and counties being involved.

The current system has homeowners getting windfalls because MERS screwed up, and I suspect that screw-ups in your version would have the opposite effect, where homeowners get robbed by computers and nobody can be held accountable.

So that seems bad. And what would the benefit be? Sure, it would make mortgage-backed securities more efficient, but -- why is that good?

...if the batch is the thing that doesn't fit in RAM, as opposed to the model.