"On the other hand this is a shitty parliamentary inquiry without legal binding what so ever."
It does have legal binding. Parliamentary committees have the power to summons people. As a foreign citizen, not resident in the country, of course, that is hard to enforce and it is not often used.
Ah yes, I had not seen that. From the faq, where is says "you can legally use..." it's not clear whether this is because the FSF don't mind, or because there is no legal ways of preventing it. Still, it seems strange to put a sign on the top of the GPL saying "modifications are not allowed" and then let it be allowed.
I am deeply confused. The first line of their comparision with AGPL shows that they have just taken the FSFs copyright of the license and claimed it as their own.
After that the license says "Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed", followed by a pile of text showing all the changes that Mongo have made.
Surely, Mongo is in violation of the the FSF copyright? You can't take someone else's work and use it as you want without their agreement?
"Meritocracy is NOT confusing. It means some people are better at tasks than others, despite their race or gender. That's it."
Okay, well that's nice a simple. Although you don't say what these tasks are. This is the problem. WIthout being clear about this, "meritocracy" is meaningless. It can mean what ever you decide it to mean.
If the task is provide a polite, comfortable environment that people are happy and free to produce their best code, then clearly by Linux is not a meritocracy because Linus is not very good at this.
This is not enough. I mean, you code might be excellent, but then it your copied it from someone else who has copyright on it, that is a bad thing. So, it's not just about code, it's about a legal and social context also.
The idea behind a code of conduct, is not to allow bad code in, but to make sure that everyone has a welcoming environment. That is, to provide a social context where all code can come in, not just some. In that sense, it is to enable a meritocracy.
"Meritocracy" is a confusing word, since it just means something where merit is important. But without a definition of merit, it's not really that helpful. Nor, is "we should just consider code that works". I mean, some who writes functional code that does the job is nice, but if that code is unreadable and unmaintainable by some else, that's not so good.
And, if we say, "he's a great coder, but an idiot whose annoys everyone in their social interactions", that is not so great. Then "merit" becomes "the ability to code and the ability to put up with an idiot who annoys everyone".
For myself, the idea that a project does not allow racial abuse, for example, is a perfectly reasonable thing to put in a code of conduct. It's not a nice thing to do, and it's not a nice thing to receive. But, more than this, in some jurisdictions, it could cause legal problems if an employer put a member of staff in a position where this happened to them and did nothing about it.
I am curious, though, where you get the idea that good code is dependent on an "unsafe space"?
We have a set of criteria rather than a list. But that is checked yes. Other places I don't know. It's not restrictive, although it nowadays the cash is starting to run out by the end of the year. WIth journals charges ranging between 400 and 4000 quid nobbling the expensive ones is an obvious way forward.
"Everyone time this gets posted, people act as though it hasn't been NIH policy for YEARS that every NIH-funded paper (so, essentially every important paper in the US) is open access within 6 months of publication.
YEARS."
Yes, thats also been true for most of Europe for YEARS. As I am sure you saw, when you read the annoucement, this is quite different. Specifically, it says that six month embargos are not okay, that hybrid journals are not okay and that article charges will be capped.
So this is quite a different thing from the NIH policy.
I think most of the points you make are inaccurate.
"In the conventional model, libraries, with professional, highly educated, librarians, basically decide which journals are legit and which are scams. The libraries negotiate bulk rates with the power of large institutions."
The libraries negotiate bulk rates with the power of an medium size institution (which is most Universities) against two or three very large publishers. In general, they get a "big deal" which means that they do not actually choose the journals. The publishers large do this.
"In the "open access" model, each individual author, many of whom see an invitation to submit to a journal in their email and assume it's legit (seriously), have to decide which journals are good, and then hand over thousands of dollars per paper. "
In the UK, open access fees are generally paid by the institution directly these days (that is a change from the past). Most often this is managed by the library, so you broadly the same people making these decisions as before.
"The vast majority of open access journals are scams."
Difficult to know whether this is true or not, but counting journals is not a good thing. There are a significant number of OA journals publishing high quality work. One big scam is the hybrid journal, where you pay for the article to be OA, but you library still pays for a subscription to the journal. The new annoucement rules this out, so this is an advance.
"Much of computer science dispenses with the extra step and just has nobody-pays journals that operate on grants at the couple-of-bucks-a-paper level."
Largely untrue also. Computing science has plenty of pay and hybrid journals, often with very high article charges. Many of these are associated with conferences, as publication in computing is largely via conference which have interesting business models in their own right.
Very unlikely that this would happen retrospectively. Also, very unlikely that the US with weak personal privacy data protections would agree to this. Especially given that data protection of this form has been around for years and years; in fact, the only place it applies in the US is when a US company is contracted to deal with data about EU citizens. This was set up to allow US companies to tender on contracts they would otherwise not be allowed to take.
Of course, not all US companies do this; they can choose to ignore the regulation but must then ignore the market. So, what you are seeing here is nothing new.
"Okay, so what you're saying is that in a world wide economics, I have to comply with often mutually excusive rules and laws."
That happens at times, and then you have to decide. For example, US tax laws mean that banks are required to release information about US citizens to the US government, even if they bank abroad, or if they live abroad. So, if you are a German bank you must release information to the US government, at least if you have any assets of US controlled territory.
But EU law (including now GDPR) says, you cannot release this information. So, you face a big fine either way.
Solution for the bank: refuse to do business with US citizens. US citizens can either: bank with a US bank (expensive if you live in Germany); renounce their US citizenship (cheap, but potentially problematic) or lobby their government to change their tax laws (good luck with that).
"I run a website with worldwide audience. I've also never been to Europe. Tell me why I should comply or face fines to a jurisdiction I've never been to?"
You don't. If you do not collect personal data, the you do not to do anything. If you do collect personal data, you can ignore GDPR, regardless of who your clients are. If you do collect personal data and part of your business is in the EU, then, yes, you need to follow GDPR.
This is not tyranny, it's just the normal process of jurisdiction.
You don't need to hire a data protection officer whatever the abstract says. You need to have some one designated in this role; it's like someone has to be responsible for safety, someone has to be responsible for first aid. A large company may well have a specific person whose job is only to do this, but most don't.
Merit means "being good at something". But being good at what?
Well, the obvious answer is good at coding, because it's a software project. So, let's make that the thing. Working code wins. Now, someone submits working code, but all comments and variables are all named in Polish, and the rest of the developers speak French.
So, let's ditch the working code wins argument; working code and the ability to speak French. We've discriminated against the Polish speaker on grounds of his inability, so that's okay, because it's all about merit.
But now, we get another developer; his code works, and hey, he can speak French. Unfortunately, he spams the mailing list all day with strange stories about his mother which are deep down a bit disturbing and off-putting. So, in the end, we ditch him, because we care about working code, the ability to speak French and not being a bit weird about your mother. And that's okay, because it's all about merit.
Finally, a developer appears who can speak French, write code, has a totally normal relationship with his mother. All good. And then 14th July happens and the developer starts shouting that Louis XVI was right all along, and offending all the other developers. So, now we have working code, French speaking, okay mother and not being offensive about French national traditions. And that's okay, because it's all about merit.
"Merit alone" is not a good argument at all. We want a meritocracy. Merit at what? If you really think that ability to code should be the only criteria then say so. Then we can have a nice quick argument which I will win in one minute. Merit alone is an simplistic not simple as possible.
The definition of organ has changed a bit over time. Originally, the organs were separable things that you can chop out with a knife. Most of them still are, but over the years the definition has been extended to include skin (which you can cut off with a knife, but is pretty diffiuse), and things like blood (which is a liquid).
Is this an organ? Well, the daily beast says that the author of the paper says that it is. But the paper that they reference does not claim this at all. And given that they are describing an anatomical feature which was "previously undiscovered", and that they are using a single microscopic technique to do this, I would say this is a bit early to claim that they found a new organ.
There are two possibilities here: the daily beast is bigging it up so that they can get a funky headline: or the author is doing this. I don't know which. The claims that this could "explain acupuncture" leaves me a bit worried that this is going to turn into another piece of pseudoscience built off one paper. Hopefully, this is not the case. We will see in a few years time.
Jaywalking has an interesting history actually. During the early days of the car manufacture, there were quite a few fatalities as a result of cars hitting pedestrians, to the point that the car industry got a lot of bad press.
So they got some heavy duty PR, invented a word "jaywalking", pushed this into the press, and lobbied intensely that it was all really the fault of the pedestrians. They managed to get it passed into law with one or two years effort. Very effective. Blame the victim.
Why do we assume that our cities should be places for cars and motor transport? Why do we assume that it's okay to drive at lethal speeds routinely? Frankly, even if you drive safely, you are still contributing to the hidden massacre caused by pollution (in the UK that's about 40,000 deaths per annum, or 10 times the numbers killed by motor traffic).
Still, it's easier to blame pedestrians, so let's stick with that rather than asking the difficult questions of why we think its okay for so many people to die every year.
Intersestingly, their article is published in a pay per view journal. One that, no doubt, is not paying the authors for their work and which, no the less, no seems to own the copyright of that article. Dystopian, working for free?
Every time rust gets mentioned, someone trots out this issue like it's the final clinching proof of, well, something. Then, shortly after, there are lots of dark mutterings about SJW.
My experience is that the rust community are quite nice. Every time I have posted an issue, someone has given me a sensible answer, explained clearly. They do have a code of conduct, but it's never bothered me because I value politeness for it's own sake. Something not that wide spread on the internet alas. The rust book is a little bit evangelic about the language, but then, that's kind of inevitable; it's written by people who have committed a lot of time and energy to a language; of course, they think it's cool and, of course, they want to sell it. That's a little tiresome but hardly unusual. On the whole, it's not bad. I don't mind a bit of enthusiasm and if someone else does the work of enforcing politeness, it's saves me the effort of blocking people client side.
And rust itself. This is an interesting one. Jim Blandy's book says, rust is all about bringing pain that other languages would give you later to the present. So, it is harder to learn and develop. Like Haskell and the ilk, there tends to be fighting with the type checker; but, then, once it compiles it tends to run. And, some designs that you might want to port from other languages do not work.
To again quote the new book, this is Rust's radical wager. It's not clear yet, whether it's a winning one or a losing one. I think it's an innovative and exciting language, but am still unconvinced that it's the right one. We will see.
I don't think you can continually water salt-tolerant plants with sea water. The salt will build up, and even salt tolerant plants cannot grow on a salt crystal.
Does this mean that all the things that broke in 17.10 that I then had to work out how to fix will now rebreak because of those same fixes? And I will have to return to where I was?
Really screwed up installing 17.10. Should have stuck to LTS.
Slashdot Mirror
"On the other hand this is a shitty parliamentary inquiry without legal binding what so ever."
It does have legal binding. Parliamentary committees have the power to summons people. As a foreign citizen, not resident in the country, of course, that is hard to enforce and it is not often used.
Actually, it's a jet and a rocket. The jet isn't fast enough.
Ah yes, I had not seen that. From the faq, where is says "you can legally use..." it's not clear whether this is because the FSF don't mind, or because there is no legal ways of preventing it. Still, it seems strange to put a sign on the top of the GPL saying "modifications are not allowed" and then let it be allowed.
I am deeply confused. The first line of their comparision with AGPL shows that they have just taken the FSFs copyright of the license and claimed it as their own.
After that the license says "Everyone is permitted to copy and distribute verbatim copies of this license document, but changing it is not allowed", followed by a pile of text showing all the changes that Mongo have made.
Surely, Mongo is in violation of the the FSF copyright? You can't take someone else's work and use it as you want without their agreement?
Yes, I agree, if there is a significant under-representation of some groups, you probably don't have a meritocracy.
Affirmative action is always a tricky one. It works in some circumstances, but can indeed cause issues.
"Meritocracy is NOT confusing. It means some people are better at tasks than others, despite their race or gender. That's it."
Okay, well that's nice a simple. Although you don't say what these tasks are. This is the problem. WIthout being clear about this, "meritocracy" is meaningless. It can mean what ever you decide it to mean.
If the task is provide a polite, comfortable environment that people are happy and free to produce their best code, then clearly by Linux is not a meritocracy because Linus is not very good at this.
"if your code is good then it's in"
This is not enough. I mean, you code might be excellent, but then it your copied it from someone else who has copyright on it, that is a bad thing. So, it's not just about code, it's about a legal and social context also.
The idea behind a code of conduct, is not to allow bad code in, but to make sure that everyone has a welcoming environment. That is, to provide a social context where all code can come in, not just some. In that sense, it is to enable a meritocracy.
"Meritocracy" is a confusing word, since it just means something where merit is important. But without a definition of merit, it's not really that helpful. Nor, is "we should just consider code that works". I mean, some who writes functional code that does the job is nice, but if that code is unreadable and unmaintainable by some else, that's not so good.
And, if we say, "he's a great coder, but an idiot whose annoys everyone in their social interactions", that is not so great. Then "merit" becomes "the ability to code and the ability to put up with an idiot who annoys everyone".
For myself, the idea that a project does not allow racial abuse, for example, is a perfectly reasonable thing to put in a code of conduct. It's not a nice thing to do, and it's not a nice thing to receive. But, more than this, in some jurisdictions, it could cause legal problems if an employer put a member of staff in a position where this happened to them and did nothing about it.
I am curious, though, where you get the idea that good code is dependent on an "unsafe space"?
We have a set of criteria rather than a list. But that is checked yes. Other places I don't know. It's not restrictive, although it nowadays the cash is starting to run out by the end of the year. WIth journals charges ranging between 400 and 4000 quid nobbling the expensive ones is an obvious way forward.
"Everyone time this gets posted, people act as though it hasn't been NIH policy for YEARS that every NIH-funded paper (so, essentially every important paper in the US) is open access within 6 months of publication.
YEARS."
Yes, thats also been true for most of Europe for YEARS. As I am sure you saw, when you read the annoucement, this is quite different. Specifically, it says that six month embargos are not okay, that hybrid journals are not okay and that article charges will be capped.
So this is quite a different thing from the NIH policy.
I think most of the points you make are inaccurate.
"In the conventional model, libraries, with professional, highly educated, librarians, basically decide which journals are legit and which are scams. The libraries negotiate bulk rates with the power of large institutions."
The libraries negotiate bulk rates with the power of an medium size institution (which is most Universities) against two or three very large publishers. In general, they get a "big deal" which means that they do not actually choose the journals. The publishers large do this.
"In the "open access" model, each individual author, many of whom see an invitation to submit to a journal in their email and assume it's legit (seriously), have to decide which journals are good, and then hand over thousands of dollars per paper. "
In the UK, open access fees are generally paid by the institution directly these days (that is a change from the past). Most often this is managed by the library, so you broadly the same people making these decisions as before.
"The vast majority of open access journals are scams."
Difficult to know whether this is true or not, but counting journals is not a good thing. There are a significant number of OA journals publishing high quality work. One big scam is the hybrid journal, where you pay for the article to be OA, but you library still pays for a subscription to the journal. The new annoucement rules this out, so this is an advance.
"Much of computer science dispenses with the extra step and just has nobody-pays journals that operate on grants at the couple-of-bucks-a-paper level."
Largely untrue also. Computing science has plenty of pay and hybrid journals, often with very high article charges. Many of these are associated with conferences, as publication in computing is largely via conference which have interesting business models in their own right.
"If push comes to shove, US companies will tell you to piss off and there's not one damn thing you can do about it."
Well, other than sue the ass of their EU assets. Which they will have because we're all global these days
Very unlikely that this would happen retrospectively. Also, very unlikely that the US with weak personal privacy data protections would agree to this. Especially given that data protection of this form has been around for years and years; in fact, the only place it applies in the US is when a US company is contracted to deal with data about EU citizens. This was set up to allow US companies to tender on contracts they would otherwise not be allowed to take.
Of course, not all US companies do this; they can choose to ignore the regulation but must then ignore the market. So, what you are seeing here is nothing new.
"Okay, so what you're saying is that in a world wide economics, I have to comply with often mutually excusive rules and laws."
That happens at times, and then you have to decide. For example, US tax laws mean that banks are required to release information about US citizens to the US government, even if they bank abroad, or if they live abroad. So, if you are a German bank you must release information to the US government, at least if you have any assets of US controlled territory.
But EU law (including now GDPR) says, you cannot release this information. So, you face a big fine either way.
Solution for the bank: refuse to do business with US citizens. US citizens can either: bank with a US bank (expensive if you live in Germany); renounce their US citizenship (cheap, but potentially problematic) or lobby their government to change their tax laws (good luck with that).
"I run a website with worldwide audience. I've also never been to Europe. Tell me why I should comply or face fines to a jurisdiction I've never been to?"
You don't. If you do not collect personal data, the you do not to do anything. If you do collect personal data, you can ignore GDPR, regardless of who your clients are. If you do collect personal data and part of your business is in the EU, then, yes, you need to follow GDPR.
This is not tyranny, it's just the normal process of jurisdiction.
You don't need to hire a data protection officer whatever the abstract says. You need to have some one designated in this role; it's like someone has to be responsible for safety, someone has to be responsible for first aid. A large company may well have a specific person whose job is only to do this, but most don't.
Merit means "being good at something". But being good at what?
Well, the obvious answer is good at coding, because it's a software project. So, let's make that the thing. Working code wins. Now, someone submits working code, but all comments and variables are all named in Polish, and the rest of the developers speak French.
So, let's ditch the working code wins argument; working code and the ability to speak French. We've discriminated against the Polish speaker on grounds of his inability, so that's okay, because it's all about merit.
But now, we get another developer; his code works, and hey, he can speak French. Unfortunately, he spams the mailing list all day with strange stories about his mother which are deep down a bit disturbing and off-putting. So, in the end, we ditch him, because we care about working code, the ability to speak French and not being a bit weird about your mother. And that's okay, because it's all about merit.
Finally, a developer appears who can speak French, write code, has a totally normal relationship with his mother. All good. And then 14th July happens and the developer starts shouting that Louis XVI was right all along, and offending all the other developers. So, now we have working code, French speaking, okay mother and not being offensive about French national traditions. And that's okay, because it's all about merit.
"Merit alone" is not a good argument at all. We want a meritocracy. Merit at what? If you really think that ability to code should be the only criteria then say so. Then we can have a nice quick argument which I will win in one minute. Merit alone is an simplistic not simple as possible.
The definition of organ has changed a bit over time. Originally, the organs were separable things that you can chop out with a knife. Most of them still are, but over the years the definition has been extended to include skin (which you can cut off with a knife, but is pretty diffiuse), and things like blood (which is a liquid).
Is this an organ? Well, the daily beast says that the author of the paper says that it is. But the paper that they reference does not claim this at all. And given that they are describing an anatomical feature which was "previously undiscovered", and that they are using a single microscopic technique to do this, I would say this is a bit early to claim that they found a new organ.
There are two possibilities here: the daily beast is bigging it up so that they can get a funky headline: or the author is doing this. I don't know which. The claims that this could "explain acupuncture" leaves me a bit worried that this is going to turn into another piece of pseudoscience built off one paper. Hopefully, this is not the case. We will see in a few years time.
Jaywalking has an interesting history actually. During the early days of the car manufacture, there were quite a few fatalities as a result of cars hitting pedestrians, to the point that the car industry got a lot of bad press.
So they got some heavy duty PR, invented a word "jaywalking", pushed this into the press, and lobbied intensely that it was all really the fault of the pedestrians. They managed to get it passed into law with one or two years effort. Very effective. Blame the victim.
The Uber video was a classic example.
Or they could drive slowly when in an area where there are likely to be pedestrians.
Why do we assume that our cities should be places for cars and motor transport? Why do we assume that it's okay to drive at lethal speeds routinely? Frankly, even if you drive safely, you are still contributing to the hidden massacre caused by pollution (in the UK that's about 40,000 deaths per annum, or 10 times the numbers killed by motor traffic).
Still, it's easier to blame pedestrians, so let's stick with that rather than asking the difficult questions of why we think its okay for so many people to die every year.
Intersestingly, their article is published in a pay per view journal. One that, no doubt, is not paying the authors for their work and which, no the less, no seems to own the copyright of that article. Dystopian, working for free?
Every time rust gets mentioned, someone trots out this issue like it's the final clinching proof of, well, something. Then, shortly after, there are lots of dark mutterings about SJW.
My experience is that the rust community are quite nice. Every time I have posted an issue, someone has given me a sensible answer, explained clearly. They do have a code of conduct, but it's never bothered me because I value politeness for it's own sake. Something not that wide spread on the internet alas. The rust book is a little bit evangelic about the language, but then, that's kind of inevitable; it's written by people who have committed a lot of time and energy to a language; of course, they think it's cool and, of course, they want to sell it. That's a little tiresome but hardly unusual. On the whole, it's not bad. I don't mind a bit of enthusiasm and if someone else does the work of enforcing politeness, it's saves me the effort of blocking people client side.
And rust itself. This is an interesting one. Jim Blandy's book says, rust is all about bringing pain that other languages would give you later to the present. So, it is harder to learn and develop. Like Haskell and the ilk, there tends to be fighting with the type checker; but, then, once it compiles it tends to run. And, some designs that you might want to port from other languages do not work.
To again quote the new book, this is Rust's radical wager. It's not clear yet, whether it's a winning one or a losing one. I think it's an innovative and exciting language, but am still unconvinced that it's the right one. We will see.
I don't think you can continually water salt-tolerant plants with sea water. The salt will build up, and even salt tolerant plants cannot grow on a salt crystal.
Does this mean that all the things that broke in 17.10 that I then had to work out how to fix will now rebreak because of those same fixes? And I will have to return to where I was?
Really screwed up installing 17.10. Should have stuck to LTS.