/dev/chat

Faheem Mitha

09:08

Can anyone recommend good places to get COVID-19 related advice? E.g. forums or telephone help lines? Do AU/UK/USA/NZ COVID-19 helplines accept international calls, and if so, are they useful and reliable resources? Also, is COVID-19 on topic somewhere on SE?

terdon

09:35

@FaheemMitha Depending on the question, both Biology and Medical Sciences. Even Bioinformatics but I doubt that's what you want.

Faheem Mitha

@terdon I just looked at Medical Sciences, but it seems advice is off-topic. Which I suppose is usual for SE.

Michael Homer

@FaheemMitha +64 9 358 5453 isn't going to know if you're actually overseas or just using an international SIM card; their advice may not be relevant to you, of course

Faheem Mitha

I need advice. I suppose I could try to couch a question in non-advice terms, but it hardly seems worth the effort.

Michael Homer

it's a generic "healthline" service

Faheem Mitha

@MichaelHomer Is that AU?

Oh, NZ. Never mind.

@MichaelHomer Yes, I see. The advice I'm looking for is fairly generic, actually.

I suppose they're reachable via Skype, though I'd also have to pay.

Michael Homer

09:46

health.govt.nz/your-health/services-and-support/…

Faheem Mitha

India has its own helpline, of course. But it would also be staffed by the Indian Govt, which is rarely a good thing.

@MichaelHomer Yes, I see. I suppose the 800 number only works inside NZ.

Michael Homer

"yes, go see your GP tomorrow" is a pretty likely response. and of course nobody is going to be calling it about any active COVID-19 situations, so they may not have anything at all useful to say

Faheem Mitha

Oh, it's not a COVID-specific helpline. Got it.

@MichaelHomer Why "of course"? Because they would be calling someone else?

Michael Homer

because nobody has had it for some time

Faheem Mitha

@MichaelHomer Ah, right. Sorry, I forget who I was talking to.

Michael Homer

09:51

they can probably manage something like "what symptoms should I look for?"; less good for "I have covid, what should I do about it?"

Faheem Mitha

Someone here tested positive for the PCR test, but she has no symptoms.

I'm wondering how to handle it. The person we've been using as a GP isn't offering the clearest advice, and getting a second opinion doesn't seem like a bad idea, anyway.

Cases here have been rising rapidly for the last few weeks, but it seems most people are asymptomatic.

Which isn't a bad thing, but it does make figuring out what to do that much harder.

Michael Homer

they are probably not equipped to help with that

Faheem Mitha

For all I know, half the city might be infected.

@MichaelHomer I guess not. I'll try the Indian help lines, and see if I can find someone who can talk sensibly. Failing that, I'll get back to the doctor.

Michael Homer

we get a lot of people arriving from India with infections

Faheem Mitha

@MichaelHomer I'm sure you do. Do they typically have symptoms?

Michael Homer

09:56

sounds like it's really ballooned there recently

Faheem Mitha

@MichaelHomer Yes, unsurprisingly. People here don't take it seriously.

Yelling also has little effect.

Michael Homer

I don't know. They wouldn't have been allowed on the plane with symptoms

But test positive on arrival or shortly afterwards

Faheem Mitha

@MichaelHomer Right. So they're probably asymptomatic. Or were when they began the transit to NZ.

terdon

What kind of advice are you after?

Faheem Mitha

My generally feeling is that Indian immune systems get more of a workout here. Like here in Bombay. So the virus had a harder time getting traction.

terdon

09:59

@FaheemMitha Does she have the option of staying home and quarantining? The vast majority of people who get covid don't develop any serious issues.

Faheem Mitha

@terdon So, one thing was that I (and my cook) were planning to get the vaccine. But the doctor said not to do that, get tested first.

Because apparently you shouldn't get the vaccine if you might have been infected.

@terdon She's doing that, as far as I know, yes. The questions weren't about her directly.

She had fever one day last week. Which is why I had her go get the test.

Though she was resistant. And it's apparently possible the test is a false positive anyway.

Michael Homer

sounds like a symptom, no?

Faheem Mitha

@MichaelHomer Sure, and she had a slight cold at the time too. But she's not had anything since then.

Adam L

@Quasímodo ok i guess i have to be pleased with any response, and dont get me wrong i am forever grateful for the selfless component of the internet that has allowed me to learn so many things without having to go back to a university and expose the local community to me in person, but i still feel like a meritocracy is an idealist delusion, having said that SE is a close enough approximation to attract people that can provide the incredibly specialized advice that i and thousands of others

have come to rely on, its just a pretty horrible feeling to hear someone say they need to "get a life" because they have 400k plus rep points in a SE community

Michael Homer

here people are classified as "recovered" when it is 1) at least two days after all symptoms have gone and 2) at least ten days since a positive test result

or at least, that was the rule a while back. I haven't kept up if it's changed

Adam L

10:06

i have about 1.5 k and that's almost entirely from the quality of the questions ive asked, im a selfish ahole i dont gaf about anyone else's struggle lol

Faheem Mitha

@MichaelHomer I see. That sounds reasonable. Though I thought it was more like 2 weeks since a positive test result.

Michael Homer

could be. two weeks after exposure is when you are considered not to have it

Faheem Mitha

I think normally they want you to retake the PCR test to make sure the virus has been eliminated.

I'm not really sure how reliable the PCR test is, though. I couldn't find any clear information or guidance about this online.

Michael Homer

however the risk management is obviously a bit different here, so I don't know what would be prudent there. your own GP/local services can presumably give you the relevant advice

@FaheemMitha genuine false positives are very rare; picking up trace amounts from people who have had it and recovered previously does happen. I would not pin hopes on a false positive.

Faheem Mitha

@MichaelHomer Agreed. Conditions in other countries may be very different.

@MichaelHomer I wasn't pinning my hopes on a false positive. Assuming it's a correct result is obviously the sensible way to go.

Anyway, thank you for the advice.

terdon

10:21

@FaheemMitha The PCR test isn't particularly reliable when it comes to negative results, but very reliable for positive. Off the top of my head, I recall a ~70% accuracy for positive (meaning 30% chance you may have it but still get a negative result).

So negative results could be considered slightly suspect, but a positive one is very clear.

The whole "recovered" thing is basically a statistics game which is why the specific threshold "two weeks after last symptom" or "one week" or whatever vary so much between countries. Many seem to agree that a week or two should be enough, but that one (how long someone is infectious for after having gone through a symptomatic phase) is very hard to quantify.

The UK NHS service has some very good pages with information: nhs.uk/conditions/coronavirus-covid-19

Faheem Mitha

@terdon I see. Do you happen to have a layman-comprehensible reference? I don't do well with biological gobbledegook.

@terdon Ok, looking at the NHS site.

terdon

@FaheemMitha For what? The test's accuracy?

Faheem Mitha

@terdon Yes, like the rate of false negatives and false positives.

Wouldn't it also depend on the quality of the lab performing the test?

terdon

I remember reading that in a respectable source, either a peer reviewed publication or a document produced by a serious clinic/research institute but I don't have the source any more. So take that with a grain of salt. I can testify to the reliability of positive results, however, since that is a characteristic of the biological techniques involved.

Faheem Mitha

Since PCR is basically an amplification technique, contaminated samples means all bets are off.

terdon

10:28

That is the same for any test. Contamination means all bets are off.

Faheem Mitha

India isn't particularly well known for keeping things clean. Though the lab in question (which happens to be next door to us), is apparently as good as it gets. For Bombay, anyway.

terdon

The thing with PCR is that you may have it, but not in enough quantities for it to be picked up in the sample. PCR is just targeted amplification of [DR]NA sequences. But you need a minimum concentration of the substrate to be amplified (the viral RNA in this case) and that is not always the case either because the infection is still in an early stage or because of sample preparation issues etc.

However, PCR is quite specific, so if it does find something, it is quite reliable.

Faheem Mitha

@terdon Yes, I understand.

A separate question. I got the impression from what the doctor said, is that if you just got the vaccine, then the PCR test might be positive, though I don't see why.

I mean, you're not dosing people with the actual virus.

terdon

Yes. The vaccine introduces traces of the virus into your system. The new ones, Moderna and Pfizer etc, which have the so-called "mRNA vaccines" actually introduce the very thing that PCR amplifies: a (non-dangerous) section of the viral RNA. So if you have had the vaccine recently, that RNA might be detected.

Faheem Mitha

@terdon Oh. Though the one here is the Oxford AZ one.

Called CovidShield or something.

terdon

10:32

@FaheemMitha You kind of are. Traditional vaccines give an "attenuated" version of the virus. So some part of the virus, often a protein that forms part of the viral envelope, that can be recognized by your immune system and so trains the immune system to react to the actual virus.

Faheem Mitha

Which I believe is being heavily used in Europe and the UK.

terdon

Yes, more so in the UK than the rest of Europe.

These new mRNA vaccines, which are really cool by the way, instead give part of the viral RNA itself. This RNA is then read by the machinery of your cell, and used to produce a tiny part of one of the viral proteins.

The actual virus would insert its RNA into your genome and hijack it to produce its proteins. The vaccine mRNA carries the instructions to build just a tiny, non-functional part of one of said proteins, so your cell will read it, produce this protein-fragment, and then your immune system recognizes it and builds immunity to it.

That way, if you ever get the actual virus, you will already be primed for it, and the mRNA given in the vaccine cannot produce the full virus (not by a long shot) so is a very safe way of priming your system.

Andras Deak

yes, <3 mRNA vaccines

They are also very effective because if the mRNA doesn't break down until it reaches your cells (see -60 degrees C storage etc.) it'll almost certainly work. Whereas you can be immune to viral vectors with those kinds of vaccines xkcd.com/2406

@terdon is that really the case?

terdon

@AndrasDeak about mRNAs? Yes

Andras Deak

Does PCR select on the spike protein, and do you get spike proteins in your nasal cavities after a shot?

sounds surprising to me

but I'm not a biology person, just musing

terdon

10:41

I don't know what exact part they target, but no it isn't the spike protein. PCR works on DNA or RNA, not on proteins. They presumably use "primers" (short DNA/RNA sequences that bind to specific other DNA/RNA sequences and then are used to amplify them) that target the viral RNA sequence, but I don't know which part of it.

Adam L

if anyone wants to be inspired by probably the most impressive unnamed nerds in history look up on youtube "UFO at JFK Airport ATC Audio" you need to listen really carefully to get a clear picture of what this person did, but trust me, no spoilers you will love it

terdon

The PCR can only answer the question "did you find traces of viral RNA in this sample"?

Andras Deak

@terdon ah right, re. protein vs RNA

so imagine that I rephrased it as "RNA encoding the spike protein"

And the RNA in the vaccines is hacked to have a fifth nucleobase. Does PCR still work on that?

terdon

They could theoretically be targeting the viral RNA sequence that encodes the spike protein but they could also just as well be targeting any unique region of the viral genome (its full complement of RNA).

Andras Deak

I'm just trying to figure out if we're talking "yeah your PCR might be positive" or "there's a mathematical chance that you get a positive PCR"

Adam L

10:43

and yes things like that need to be illegal because it would literally cause a mass stupid person psychosis and we would have a mad max environment

terdon

@AndrasDeak I don't know about that I would need to understand what this 5th base is. Are you sure?

Andras Deak

absolutely, hold on

in Python on Stack Overflow Chat, Dec 28 '20 at 10:50, by MisterMiyagi

Random reading of the day: Understanding RNA as string/instruction encodings if you already know the absolute minimum about unicode and character sets.

read the first link

but I thought that was linked here too and we discussed it

terdon

Ah. Neat, I hadn't seen the bit about the modified U. But yes, the relevant part is:

> The really clever bit is that although this replacement Ψ placates (calms) our immune system, it is accepted as a normal U by relevant parts of the cell.

This is new to me though. I don't know what difference it will make, so I have no opinion on it at all.

Andras Deak

I'm curious because there's a similar topic of vaccinating lactating mothers. The vaccines haven't been tested yet so the rigorous stance is "we don't know if it's safe".

But the CDC says "We don't know, but the artificial mRNA is unstable so it's unlikely to go far from the arm and probably won't reach the mammaries, if it reaches the mammaries it probably won't be excreted in breast milk, and if it does get excreted it will probably be digested by the baby. And the potential benefits of maternal antibodies trump this mathematical chance of problems".

So I'm wondering if we're talking realistic chances or mathematical ones with the PCR.

Faheem Mitha

10:59

@terdon The question of which bits the PCR targets is obviously relevant.

If I understand you correctly, the issue is that if the PCR amplifies the bits of DNA/RNA in the vaccine, it will confuse the test.

Andras Deak

since people seem aware of the specific strain that's found in people, I'd assume that PCR looks for the entire genome

Faheem Mitha

So the bottom line is that being vaccinated can invalidate the test, but the probability of invalidation is indeterminate?

Andras Deak

Well that's a fairly safe bottom line. The question is always the probability.

Faheem Mitha

Hmm. Well, that's not ideal. But perhaps unavoidable.

Andras Deak

We are a bunch of laymen musing about something technical. One could see what people familiar with the intricacies of covid PCRs have to say. Otherwise this is all academical navel-gazing.

Faheem Mitha

11:14

@AndrasDeak Where would one find such people?

Might be on-topic on Medical SE, I suppose. Assuming it hasn't been covered already.

Andras Deak

I wouldn't be surprised if this was already available information on the internet. Our sources have so far been 1. one doctor you talked to, 2. whatever terdon and I think, 3. an article about reverse-engineering the pfizer vaccine

for instance the CDC and other offices might already have FAQs along these lines

Faheem Mitha

@AndrasDeak A general health site isn't going to have a technical discussion.

But sure, it's likely the information is available online somewhere.

Whether any of us would understand it is another matter.

So is this mRNA technique fairly new then? Sounds high-tech. Not exactly Edward Jenner.

Andras Deak

@FaheemMitha very new

Stephen Kitt

the techniques have been in development since the 90s, but the COVID mRNA vaccines are the first to be approved

Andras Deak

In large part thanks to a Hungarian scientist (Katalin Karikó) who was driven away with her research that was deemed useless here. She was forced to go abroad, then she was driven away from academia. She ended up using her/their mRNA research to contribute to this new kind of vaccine.

Stephen Kitt

11:24

@FaheemMitha you might be surprised, see for example cdc.gov/coronavirus/2019-ncov/vaccines/different-vaccines/…

Andras Deak

She always makes it clear that this was a group effort, very humble lady. Also brilliant.

Faheem Mitha

@AndrasDeak Yes, I was just looking at her Wikipedia page.

Andras Deak

and of course she's a scientist so it wouldn't have come together without the bioNtech industry side

I think there's a tweet somewhere by Richard Dawkins demanding a Nobel for her and coworker :)

Faheem Mitha

@StephenKitt Sure, but the question was (I think) what bits the PCR process chooses to amplify, and whether it would coincide with what the vaccines have.

terdon

@AndrasDeak Nah, it will be targeting a specific part of it.

Faheem Mitha

11:30

@AndrasDeak Science can be a rough business. They don't tell you that in the brochure.

Google threw up statnews.com/2020/11/10/…

terdon

@AndrasDeak For what it's worth, I am not a layman on PCR, RNA, and genomics. I am not a medical doctor, and am at best a mildly informed layman when it comes to the clinical details, but I do have a very deep understanding of how DNA works. I should do! I've got a PhD in genomics and have been working in the field for 20 years, after all.

Faheem Mitha

Which is light on the science, so I can understand it. But would likely not satisfy a biologist.

terdon

Anything I say about the medical effect, covid etc, should be treated as the noise of a navel gazing layman though, yes.

Andras Deak

@terdon I wasn't trying to imply that, I know you have some kind of bio/medical history. But you seemed mostly unsure on the subject :)

terdon

Indeed I am.

Faheem Mitha

11:32

I spent years in what was (supposedly) a immunology research group, but never really got a handle on the biology. Though I'm told it is "easy". For some value of easy, I suppose.

terdon

snort. The same value of "easy" that you would apply to math :)

Andras Deak

It's easy for the kind of people who find biology easy.

Faheem Mitha

@terdon @AndrasDeak Indeed, yes.

Reading biology papers either give me a headache, or put me to sleep.

The acronyms by themselves are bewildering.

terdon

Yeah, that's how I feel about math papers. Luckily, I've very rarely needed to read any.

Andras Deak

CDC and first page of duckduckgo hits say you won't test positive with PCR after vaccination

Faheem Mitha

11:34

Math is easy in comparison. At least they don't have multiple polysyllabic names for everything.

Andras Deak

cdc.gov/coronavirus/2019-ncov/vaccines/facts.html

> After getting a COVID-19 vaccine, will I test positive for COVID-19 on a viral test?

No. Neither the recently authorized and recommended vaccines nor the other COVID-19 vaccines currently in clinical trials in the United States can cause you to test positive on viral tests, which are used to see if you have a current infection.

(and US Moderna is mRNA)

Faheem Mitha

@AndrasDeak Interesting. It would be nice if they could offer some justification.

But perhaps they aren't covering Oxford?

Andras Deak

I suggest perusing duckduckgo.com/…

Faheem Mitha

If Moderna and Pfizer use this mRNA technique, that would explain why they are so effective.

Andras Deak

yes

as long as no mutation affects the spike protein too much :)

Faheem Mitha

11:39

@AndrasDeak Right.

terdon

@AndrasDeak I wonder if that is a question of the specific tests in use in the US.

Andras Deak

@terdon it probably is. But I didn't realize there was a difference.

terdon

I am making this up as I go along. I'm thinking that if the PCR-based tests available in the US are targeting something they know is not part of the vaccine mRNA that would explain it.

Faheem Mitha

So can the PCR tests target specific fragments? That's the first question.

Andras Deak

@FaheemMitha yes

Faheem Mitha

11:42

And if so, is that then hard-coded into the test?

Andras Deak

they look for strands of DNA (or RNA with a trick I think)

you can choose what to look for

Faheem Mitha

So it's more like a class of tests, then. Depending on what is being searched for, exactly.

Andras Deak

Yes, PCR is just a technique that amplifies whatever DNA is in a given sample. The acronym stands for polynuclease chain reaction, which is just a chemical reaction that generates more DNA from little DNA. As I understand at least. Terdon can smite me if I misspeak.

terdon

@FaheemMitha Yes. It depends on the "primer sequences" used. These are pairs of short, unique sequences that are complimentary to a specific RNA sequence. Because they are complimentary to it, that means that they can bind to it and then they act as templates for amplification. You use two primers, one for the beginning of the target sequence and one from the end, and they result in amplifying the sequence between them.

@AndrasDeak no smiting! :) Polymerase Chain Reaction, but that's pretty close!

Andras Deak

bah, I knew that

Faheem Mitha

11:53

@terdon So one would do the test with the supplied primer, or primers. Which is kind of like a reagent?

As usual, I'm not sure what any of these words mean. Sigh.

Andras Deak

Sounds like choosing the right kind of soil to cultivate what plant you want among all the seeds. Except this is completely specific, because no other RNA will bind to a given half-chain

terdon

@FaheemMitha Yes, it is a reagent: think of a tube with liquid in it.

Stephen Kitt

12:10

or regexes to define start and end patterns in AWK or sed

Faheem Mitha

12:23

@terdon OK. I can handle a tube with liquid in it.

Handle mentally, I mean.

terdon

@StephenKitt It's very close to regexes actually! Oooh, hang on, you guys might like this. The CTO of the company I work for found a really cool blog post called "genomics for programmers", or something like that, which explains various genetic/biological concepts in computer science terms. It has really helped a lot of our programmers grok what the hell it is we do here. Let me see if I can find it.

Stephen Kitt

@terdon yes, my brother works in genomics, he sometimes mentions new ways he’s found of applying Unix-style tools to ACGT/ACGU sequences

but nowadays he mostly uses R

terdon

@StephenKitt Sigh. Everyone does these days and I don't know why. It is an awful language, very ill suited to the task but there are now so many great tools written in it that everybody uses a screwdriver to hammer a nail.

berthub.eu/articles/posts/amazing-dna

That's the one I was thinking of if I remember correctly.

Faheem Mitha

@terdon At least that's one thing we agree on. R's awfulness, I mean.

I remember when I started using it, around 1998/1999, I felt differently. It all seemed very glamorous. But one gets jaded, unfortunately.

For example, ggplot2 is amazing.

Jeff Schaller

@StephenKitt R, the newly-discovered nucleotide‽

Stephen Kitt

12:37

@JeffSchaller R, because it’s not as good as all the stats tools with names starting with S

Jeff Schaller

@StephenKitt I heard that the authors of R ended up pirating the S software

terdon

@FaheemMitha a lot of things in R are amazing. It's just such a pain to use. When you get it working, the results are great.

Faheem Mitha

@JeffSchaller They didn't. It's a new implementation.

Even the language is different. It's very close to S, of course.

The language S, that is.

Written by Chambers. If I recall correctly.

Jeff Schaller

@FaheemMitha they did; they pronounce it arrr

often followed by "matey"

Faheem Mitha

@JeffSchaller Huh?

Stephen Kitt

12:39

think “talk like a pirate day”

Faheem Mitha

@StephenKitt I realise pirate references are being made. I just don't follow the relevance.

And even the authors of R (at least one), admit they didn't know what they were doing.

Jeff Schaller

@FaheemMitha my typical bad pun, sorry -- pirate: the verb versus pirate: the noun

Faheem Mitha

And it shows.

@JeffSchaller OK.

I felt positive about R once. I felt positive about C++ once too. But one learns and evolves.

terdon

Well, at least R is open source and free.

Faheem Mitha

13:33

So, to summarize. Per my understanding, and according to information received, someone taking the vaccine may or may not have a positive PCR response, depending on what reagents have been used, and if they have been chosen to avoid the DNA/RNA fragments that were present in the vaccine?

@terdon It's Free as in Freedom.

terdon

13:48

And as in beer.

Faheem Mitha

@terdon That follows from Free as In Freedom

terdon

@FaheemMitha No it doesn't! You can have sell software that's Free as in Freedom: you share the source code, and people are free to modify and redistribute.

Andras Deak

@FaheemMitha and whether the mRNA can reach one's nasal tissue

Faheem Mitha

@terdon The rules of Free Software don't permit restrictions on redistribution, so yes, it does.

Of course, there are variants of the rules, but they all amount to the same thing.

Freedom 2 in gnu.org/philosophy/free-sw.en.html

That's the FSF version, of course, but Debian and OSI for example, are no different.

Stephen Kitt

@FaheemMitha no, it doesn’t; the fact that it’s possible doesn’t mean that it’s inevitable. DFSG-free software isn’t necessarily 0-cost.

Faheem Mitha

13:56

@StephenKitt Yes, it's inevitable. Assume you distribute it in the first place, of course.

But if you don't, the free software rules don't come into play. I guess one would call them a dead letter.

Stephen Kitt

@FaheemMitha no, it’s not inevitable. You need someone somewhere to make it available for free.

Faheem Mitha

@StephenKitt Huh? I don't understand what you mean.

ilkkachu

He means exactly what he said, I think

Stephen Kitt

@FaheemMitha I can release a project under the GPL, but only distribute it to people who pay me for it. It only becomes available for no cost if one of those people then turns around and distributes it for no cost.

The no-cost availability of the project is not inevitable.

ilkkachu

Case in point, grsecurity.

Faheem Mitha

13:59

@StephenKitt Right, but the only reason it wouldn't eventually be distributed for free is if restrictions were put on it.

@ilkkachu grsecurity?

Stephen Kitt

@FaheemMitha no.

ilkkachu

See e.g. lwn.net/Articles/721848 and grsecurity.net/download

@FaheemMitha that doesn't follow, unless you make assumptions on the behaviour (and motives) of those paying customers.

Faheem Mitha

@ilkkachu So, if I understand correctly, those paying customers could give those patches away legally. Per the rules of the GPL.

Assuming anyone cared of course.

ilkkachu

@FaheemMitha yes. I don't think anyone was arguing they couldn't.

@FaheemMitha also assuming they want to go against the wishes of their source.

Andras Deak

I think the underlying confusion here is that Faheem takes it for granted that if one person pays for the code then they will put it somewhere online because they are allowed to, and henceforth anyone else can get it for free.

unless I'm misunderstanding and only contributing to the confusion

Faheem Mitha

14:07

@ilkkachu Wishes? There were wishes?

@AndrasDeak No, but in practice if free software is distributed to a significant numbers of free customers without further restrictions, if the software is of interest to others, it will inevitably be further distributed. I'm not taking into account "wishes", of course.

Which sound like a contravention of the GPL, in spirit, if not the letter.

And it's very unlikely the people further distributing it would charge for it. I suppose anything is possible, though.

ilkkachu

@FaheemMitha Sigh. Nuances. If company A provides a GPL:d software for its customers in return for money, and doesn't explicitly distribute it for free to others, then don't you think there just might be some kind of an implicit wish from their part that the customers also don't distribute it for free?

Andras Deak

hehe, Finn spotted

Faheem Mitha

@ilkkachu Free software, per its definition, doesn't put restrictions on distribution.

ilkkachu

@FaheemMitha where did I say anything about restrictions?

Faheem Mitha

I really don't know what part "implicit wish"es would play in this.

@ilkkachu Your "implicit wish" sounds like an "implicit" restriction. :-)

That would actually be a reasonable question to ask on Open Source SE.

Is people wishing stuff OK per the rules of the GPL, say.

Andras Deak

14:12

There's an open source SE? Sounds scary.

Faheem Mitha

@AndrasDeak Yes. Why it is scary?

Andras Deak

Well, this discussion right here, when extrapolated to a whole SE site, is scary.

Faheem Mitha

@AndrasDeak You have interesting notions of scary.

Andras Deak

I didn't want to use less welcoming terms

Faheem Mitha

I'm mildly tempted to write to Richard Stallman and see if he replies.

I exchanged two emails with him once, but they weren't anything interesting.

He's very polite over email, at least.

ilkkachu

14:14

@FaheemMitha So, how do I put this so that's clear enough but not too blunt. Honestly, I think what I said reads like that if one reads what they want to read, instead of what was written.

Or includes some strong assumptions in their viewpoint.

Faheem Mitha

@AndrasDeak Don't feel you have to spare my feelings. It's not my site.

@ilkkachu Sorry, I can't make my understanding other than it is. :-)

ilkkachu

I don't know about others, but I've been taught that there's some difference between "a wish" (or "request"), and "a requirement" (or "restriction", "demand", whatever)

But in any case.

Faheem Mitha

@ilkkachu Sure it is. But my issue is whether possibly behavior modifying "wishes" are OK, per the GPL. Does it constitute a restriction, or not?

Andras Deak

RMS indeed sounds like the ideal person for this question

ilkkachu

@FaheemMitha well, does it?

Faheem Mitha

14:16

It seems to me that if it was a verbal "wish", it would not be legally enforceable. But it also wouldn't propagate very well.

ilkkachu

"I'd be happy if you didn't do this", vs. "You can't do this"

Faheem Mitha

@ilkkachu I don't know.

Andras Deak

@FaheemMitha verbal contracts are contracts

just saying

Faheem Mitha

@AndrasDeak Then it's in violation of the GPL. Assuming the GPL is the contract being used.

But I'm not a lawyer, of course.

Andras Deak

and we're not talking about contracts

Faheem Mitha

14:18

@AndrasDeak We aren't? The GPL is a contract, isn't it?

Andras Deak

I just wanted to point out that the 'verbal' in 'verbal "wish"' is irrelevant

ilkkachu

@FaheemMitha a license, no?

Andras Deak

nevermind, I remembered why I didn't want to engage in this

Faheem Mitha

@AndrasDeak Actually, it's all important. Because if it was written down, then it would become something that had official legal weight. And would violate the GPL.

ilkkachu

@FaheemMitha "The GPL was designed as a license, rather than a contract." -- en.wikipedia.org/wiki/GNU_General_Public_License (there's probably a better source)

@FaheemMitha and if it isn't?

Faheem Mitha

14:19

I'm speaking loosely, but what I mean to say is that you could then prove that there was intent to violate the GPL. A verbal request or wish would be "under the table", so to speak.

@ilkkachu See my followup.

Stephen Kitt

Verbal agreements are legally binding in many jurisdictions.

Faheem Mitha

@ilkkachu Again, not a lawyer, but isn't a license a kind of contract?

@StephenKitt I've always wondered about that. How do you prove they were made?

Stephen Kitt

@FaheemMitha no, because a contract requires the possibility of negociation.

ilkkachu

GNU General Public License

The GNU General Public License (GNU GPL or simply GPL) is a series of widely-used free software licenses that guarantee end users the freedom to run, study, share, and modify the software. The licenses were originally written by Richard Stallman, founder of the Free Software Foundation (FSF), for the GNU Project, and grant the recipients of a computer program the rights of the Free Software Definition. The GPL series are all copyleft licenses, which means that any derivative work must be distributed under the same or equivalent license terms. This is in distinction to permissive software licenses...

Stephen Kitt

@FaheemMitha witnesses, or a recording, for example.

Faheem Mitha

14:21

@StephenKitt Oh. Yes, I see.

Well, dinner time. Maybe I really will write to Stallman. I hear he answers his email. An opportunity to find out if that is actually true.

ilkkachu

@FaheemMitha, anyway. People sometimes have other motivations than doing what they can. E.g. keeping good relations with their software provider.

Faheem Mitha

@ilkkachu Yes, of course.

Andras Deak

@FaheemMitha don't forget to ask him about what he thinks of the controversy surrounding his return

Stephen Kitt

A license isn’t a contract because it’s a “take it or leave it” proposition. Note too that for example the GPL does not require that users accept it in order to use the software covered by it, assuming they obtained it properly.

Faheem Mitha

@AndrasDeak I'll make sure not to ask him that.

ilkkachu

14:22

But yeah, I wouldn't be unhappy if someone were to redistribute the grsecurity patches in public, for free.

Faheem Mitha

@StephenKitt Agreed. It only comes into force when distributing. Of course.

ilkkachu

(I haven't exactly looked if that has already happened within the past years, but any case.)

Partly because I would want to see the current patches, partly because I would want to see the followup to that happening.

So yeah, please, go give them money and then put the patches out for free.

Faheem Mitha

Is anyone aware of any other examples? Because if there were, I could mention them.

Stephen Kitt

Quite a lot of the interesting grsec stuff is part of the mainline kernel now anyway...

terdon

@FaheemMitha the main point is different: if I make a tool, and I am responsible for it, know its code and am working on it, I can then distribute it for a price, only to paying customers and as long as I don't REQUIRE them to not distribute it again, even if I ask them nicely not to, then that tool can be GPLed.

Now, I can also make a valid business model out of this because although everyone is free to take my tool, modify it and redistribute it for free, my customers get two things: i) the seal of quality offered by the author, lets assume I'm some sort of expert in my field and can produce software of excellent quality because of my expertise and ii) the support offered by the author which the redistributers cannot do.

More importantly, since I am an expert in this field, and let's also assume I'm a great coder, then we can also assume that very few people in the world will have the required domain knowledge and programming chops to duplicate my work and it is cheaper for my clients to pay me to get the code than to try and figure it out themselves. Plus, since the code is GPLed, any improvements made can easily be put back into my original work and I can contuinue selling it.

ilkkachu

14:29

Plus if you don't like what your clients are doing, you can always choose not the renew their subscriptions, or sell them the next version or whatever.

@StephenKitt I know there's progress on that, but I haven't followed too closely. And I got the impression earlier that it was kind of hard to get pre-emptive protections in past Linus.

Stephen Kitt

@ilkkachu things are better on the latter front now; see github.com/KSPP/linux/issues and kernsec.org/wiki/index.php/Kernel_Self_Protection_Project

ilkkachu

yep

Faheem Mitha

16:51

@terdon I suppose that's a possible model. But I've not heard of anyone doing that. Of course, if it was closely held code, then I suppose I wouldn't.

At any rate, I've not heard of this happening myself, ever.

@terdon As mentioned earlier "ask them nicely not to" might be a GPL violation, btu again, not a lawyer.

terdon

@FaheemMitha It's basically what people like Redhat and Canonical do, the part about selling support anyway, granted not the part about selling code. But I do remember having paid money to buy some CD burning tool, norton something?, many years ago and that was opensourced.

Faheem Mitha

@terdon People can and do distribute RH code. I think the only requirement is stripping out trademarks? Isn't that essentially what CentOS was doing?

terdon

@FaheemMitha Yeah. I read the transcript. Not a lawyer either, but I am pretty sure that is a very flimsy argument. If "please don't distribute" counts as a restriction on distribution rights, then "go jump of a bridge" is murder.

Faheem Mitha

I don't know if CentOS still exists. I've never really used it.

@terdon I guess that's a fair point.

terdon

In any case, the point is that "libre" software, does not need to be free of cost.

It just usually is so we're used to it.

Faheem Mitha

16:55

@terdon Still wondering if there are any actual examples of this hypothetical scenario.

Not that it really matters.

@terdon There are very powerful forces pushing in that direction. For one thing, every coder on the planet who isn't actually working for hire wants his code distributed as widely as possible, so that people can see it, and hopefully improve it. Or at least report bugs.

And also, for the economic model you described, a dual licensing system is a standard approach. E.g. MySQL used to do that, I think.

Well, it's not your model exactly, but it's a model whereby one can actually sell GPLd software.

terdon

@FaheemMitha that isn't relevant. We're talking about someone who is doing it to get paid.

Here, @FaheemMitha, take it from the horse's mouth: gnu.org/licenses/gpl-faq.html#DoesTheGPLAllowMoney

> Does the GPL allow me to sell copies of the program for money? (#DoesTheGPLAllowMoney)
Yes, the GPL allows everyone to do this. The right to sell copies is part of the definition of free software. Except in one special situation, there is no limit on what price you can charge. (The one exception is the required written offer to provide source code that must accompany binary-only release.)

Faheem Mitha

@terdon I understand that the person is doing it to get paid. But even people who are being paid for their software like other people to look at it.

@terdon That one can sell copies of the program for money isn't in dispute. The GPL clearly allows that.

The issue was about further redistribution by the recipients.

terdon

@FaheemMitha it was?

3 hours ago, by terdon

@FaheemMitha No it doesn't! You can have sell software that's Free as in Freedom: you share the source code, and people are free to modify and redistribute.

3 hours ago, by Faheem Mitha

@terdon The rules of Free Software don't permit restrictions on redistribution, so yes, it does.

Faheem Mitha

@terdon Sure. I quoted gnu.org/philosophy/free-sw.en.html

terdon

And everything since then has been an attempt to convince you that yes, you can sell free software for money.

Faheem Mitha

17:09

@terdon Yes, that's what I wrote. You can't put restrictions on redistribution. But you can still sell your software. There is no contradiction.

You just can't control what the people you sell it to, do with it. That's all.

terdon

Well no, of course not.

@FaheemMitha Sigh.

Faheem Mitha

And yes, maybe they will choose not to redistribute it.

terdon

You claimed that "free as in beer" (i.e. gratis, for free, no money) "follows from Free as In Freedom".

That's what we've been discussing for three hours!

jesse_b

where do I inquire about this free beer?

terdon

@jesse_b Free beer is different. You may have it, but redistribution is only permitted in your own bathroom, buddy.

jesse_b

17:14

yes but where do I get it

terdon

@jesse_b free-beer.co.uk

jesse_b

hey that's not free beer, that is beer free

terdon

Yeah. Bastards!

freebeer.org/blog

better?

> FREE BEER version 6.0 (the Atlantic brew) is an open source beer based on classic ale brewing traditions but with “added real ideas” from RIO.

Damn, that's geeky.

jesse_b

Beer has become pretty geeky

terdon

True. But I'm glad of it, actually, since I never liked beer until I discovered micro-breweries and non-lager beers. Turns out I don't really like lager, is all. As I recall, you're lucky enough to prefer lager, so you never had to go through the lean, beerless years of my youth :P

jesse_b

17:26

I was just going to go on a rant about microbreweries

the gall of people to think they can do it better than multi billion dollar companies

if your microbrewery was any good it wouldn't be micro for very long

terdon

ha! If your company were doing anything worthwhile, and spending the money you need to spend on decent ingredients, you wouldn't be worth billions!

That's like saying "If your italian granny could cook worth a damn, she'd be rich!" :P

jesse_b

touche

well not really though because the italian granny isn't selling her cooking

The microbrewery is making beer with the sole purpose of profits

ilkkachu

@FaheemMitha so... was there something in grsecurity that didn't fit as an example?

Adam L

so if i go to the psychology chat rooms am i allowed to talk about how insane it is that Peterson feels so compelled to give the mass life advice and even worse, take nutritional advice from his psycho daughter?

jesse_b

Adam L

17:39

im just saying, they are terrible representatives of the conservative view they need new reps

jesse_b

Well I still don't know what you are trying to say but aside from that why are you saying it here?

Adam L

thats why he needs to supplement his career with family tragedy and a public dialog with his psycho but attractive daughter

terdon

@jesse_b Philistine! They're doing it for the pleasure of the craft and the shine it puts into their oh so luxurious yet carefully coiffed beards!

Adam L

because you guys are less nimrod than any other chat room the math rooms have banned me for 90 days for uber tame crap

jesse_b

@terdon if all the cared about was the pleasure of the craft than it wouldn't be so hard to find free beer!

terdon

17:42

@AdamL Maybe you should learn from that and stop testing the waters here, then?

@jesse_b Dude. You think that cruelty free, organic, vegan eyebrow conditioners grow on trees?

jesse_b

loled

@terdon: how did your work issue go yesterday? You get all those things rechecked?

Adam L

hey enough food talk someone stole my egg plant from my shopping basket and it isnt even funny in reference to the emoji culture on chat apps now i have to wait till 8 am

terdon

@jesse_b Still on it. But I now know that the vast majority were barely affected at all, so I can sleep a bit better at night. Yay!

Adam L

but point taken i shall depart in receipt of the implied threat of ban

2

jesse_b

@terdon did you automate the validation?

terdon

17:48

semi... I ended up writing various little scripts

All I really need to do is compare the old and the new and see if the new has things the old had missed. But since these are >900 analyses, which have run with different versions of various tools, it became a bit of a nightmare. Almost done now though. All that's left is the fun part: informing clients that we had an issue.

jesse_b

heh

terdon

In other news, anyone have any idea how the OP can have keybindings that don't appear in dconf? I really doubt they've setup their own xbinkeys or anything like that:

1

Q: Linux Mint 20.1 unbind "hidden" shortcuts

Linux Mint 20.1 with Cinnamon opens the terminal when hitting Super+2 and brings windows to the front with Super+3 and Super+4. Super meaning the Windows-Key. I would like to use these shortcuts otherwise, but going into the shortcuts tab of the "keyboard" window, the unwanted shortcuts are not e...

Andras Deak

@terdon ~~I feel censored~~ (censored)

Transcript for