Statistically people without suitable photo ID tend to be minorities.
Except that everyone with the right to vote is in elebenty-one government databases. You had a birth certificate issued, you have a SSN, you pay taxes, in the US you can't seem to get your groceries without a car, etc. So there's more than enough ways you can point to existing government lists -- unless you're there illegally.
Which is a good part of why Democrites have their panties in a twist. Repugs have a point here.
Yeah, the breakage is not your fault. That timestamp you included might still be encoded somewhere inside the massive URL that sign-away-your-firstborn page redirects to, getting past it is currently beyond my threshold of putting up with crap.
Google somehow insists that everyone with an IP in country X prefers language X, even when accessing a country-Y-specific URL, with browser sending English as the preferred language. Searches in non-English are mostly useless except for some local stuff -- but I for one never shop/etc using Google. For plain searches there's DuckDuckGo, but there's far more to Google than that.
For a more extreme horror show, try Google over Tor. Their captchas are broken most over time.
Your link is broken -- it redirects to a country-specific YouTube site that dims the page and presents an empty white rectangle. If I recall correctly, if I dropped enough privacy-related extensions, it'd show me a form to sign away most of my rights, then keep doing so unless it's allowed setting a 3rd-party cookie. Just no.
Redirecting someone to youtube.pl when following a link to youtube.com is inexcusable, as content of these two is different (even if a good part of videos can be watched on both). They used to have an opt-out but AFAIK it's gone with no replacement.
I also have no 1h7m to waste to watch a talk I could read in 5mins or skim over in much less.
The problem is that there is not one RFC that defines an email address!! There is one that covers the top level domain (if present), one that covers the domain name (if present), and one that covers the account portion.
5322 is quite informative. The only simplification it does that I see on the first glance is that it defines "domain" as "dot-atom / domain-literal / obs-domain", dot-atom being a series of 1 or more atoms [a-z0-9!#$%&'*+/=?^_`{|}~-]+ separated by dots (ie, there might be no dots, but a dot can't be the first or last, and two dots can't come in a row). This allows using non-Internet domains, but if you know you'll use only them (a quite safe assumption), you know to refer to the DNS RFC. I'd risk saying that a policy that bans domain-literal is reasonable.
There isn't even a mandate for the @ to be present as it is not needed when sending email within one domain.
Yeah but a domain-less mail is useless on a contact form. It's about same as you said "exit our office, go left, the cafe is on the corner" vs "Sesame Street 42, Springfield TX, USA" -- valid only in a very limited number of contexts.
What makes it act like a standard is that everyone adheres to it so practically it doesn't matter if it is stuck in RFC status for eons.
e-mail RFCs are actually updated relatively often to keep being relevant, although they like to keep compat baggage.
Depends on whether you want to support pre-Internet addresses, as that's a good part of the spec. If you drop stuff like routed addresses or comments/whitespace inside a domain, the rest is quite simple. Your MTA likely won't accept "@node.test:mary@ (this is a comment) example (whitespace around.).net" thus it should be dropped at validation time.
Actually, most libraries are wrong, too. If you look at nodejs, it's astonishing how 90% of packages that ship a single line of code (plus tons of boilerplate) get even that single line wrong.
It doesn't take a genius to look up the relevant RFC and write a regexp.
I'll add to that that this isn't limited to web developers.
Hell yeah this. In most fields, there's decades of prior wisdom that one should at least try to get a grasp of before making "brilliant" inventions. Like, you shouldn't write an init system before reading "Unix for Dummies".
Things like SIGHUP/nohup are basic knowledge, if you don't know this you shouldn't write your snowflake way of killing processes on logout. If you don't know ways to authenticate users and what user names are explicitly allowed by POSIX (the 0day issue), you shouldn't create a gaping security hole, and even worse, you don't WONTFIX it just because one of your distribution's clicky-clicky tools doesn't allow such names, while "$EDITOR/etc/passwd" or LDAP are "user errors".
I'm for one a kernel newbie -- yet I know better than when faced with a problem like "shit network driver does high-order allocations while atomic, with disastrous reasons when under memory pressure" to rush into making an ad-hoc pool instead of reading how it should be done, or asking those with a clue. The former was my first reaction, yet I at least bothered to think before wasting developers' time with a bogus patch.
But the vast majority of men don't choose to become engineers either. So overall, it's more a matter of choice.
And here you nailed the most important factor.
But, when voting for a politician or hiring an employee, you don't get a time machine to their kindergarten to tell toddlers that a toy car fits girls while it's not a shame for a boy to play with dolls. No matter how hard you try to change early stages of the pipeline, any effects will be visible after more years than your company even exists. You get to pick from the pool of candidates available today.
I don't believe that's because women are any less capable
All observations show that they're indeed less capable here -- on the average. And that's not surprising: there's so many physical gender difference, such as body water content ratio, brain sizes, muscle and fat layout, ability to discern colors and smells, etc -- thus, assuming there are no mental differences would be preposterous. But, all that average tells you is that, with a gender-blind criteria, one group will make a smaller part of the population than the general count would say.
Personal variance trumps gender/race differences, thus, with a distribution without a hard cap, there will be skilled individuals from both groups above any threshold (assuming an infinitely large population).
Example of a distribution with a hard cap: D&D-like stats. Orcs get int of 3d6-2 str 3d6+2, elves 3d6+2 str 3d6-2. Thus, all the top wizards will be elves with ints 19 and 20, with not a single orc above 16. Likewise, all top fighters will be orcs. But in real life, the distribution has no such cap (ok, ok, physics tells you you can put only a certain number of processing elements and mechanical strength into a given volume, but human biology is so many orders of magnitude below physical limits that this is akin to arguing that a "no limit" stretch of Autobahn is still subject to the speed of light).
Thus: a woman picked at random is likely to be less strong than a randomly chosen man. But I still wouldn't want to get into a fight with a female lumberjack. Yet you're not going to argue that women are as capable at lumberjacking on the average, and it's only evil woodcutter companies having a bias towards men.
Oh sweet Summer Child. You actually believe Open Source is immune to politics and is a perfectly level playing field. How endearingly naive.
Well, the field obviously isn't level. Outreachy, Debconf travel funds, etc -- people indeed do get discriminated against based on their gender.
Female contributors just aren't there. I for one do quite a bit of mentoring -- out of 137 packages, there's just one upload by a woman. It was a fine upload, perfect on the first try while most people need multiple attempts. But it's the only one.
I try hard to not discriminate or even say things in a condescending tone, and I got the impression I'm doing ok -- as once the chaff is filtered out (I don't even see those who don't try to submit a package), there's no gender difference. In this case, on above corpus, women package quality is even drastically higher, but with just a single data point, this is not statistically significant.
Thus, women who do submit kernel patches or Debian packages are no worse, at all, from their male peers. The problem is, they make such a tiny portion of the submitters.
Things change once they get paid for coming. I don't believe skills of paid employees are drastically different from unpaid volunteers, thus the skew is obvious.
And what about the other tech companies, who also sport a similar overabundance? They sing in one voice here.
Which is greatly harmful to the women who actually are skilled, as they indeed tend to get dismissed by their peers as mere diversity hires. Which spins up the wheel of hate even more.
Let's compare: among 1000 kernel devs with most commits who have names that show gender (I know western and slavic naming conventions), there's 8 women. In other similar projects, gender ratios seem to be similar. Yet for example Google has 26% women among engineers. As open source projects are driven by work you actually do rather than by irrational hiring decisions, I believe it's the former figure that's representative of skill (or rather, of who decides to learn these skills). Thus, only possibilities are that 1. Google picks employees based on their race/gender rather than merit, or 2. there's a glut of skilled women somehow denied employment elsewhere. As most other big tech companies also tend to be SJW-run, they have an overabundance of diversity hires, thus we know 2. is not the case. Thus, a good part of women employed have worse skills than their peers. And thus, either they get paid less, or get unfair wage increases.
This is not to say that very highly skilled female/minority engineers don't exist -- they do, I can point you to a few. The thing is, they are rare.
This ban is for indoor. I could only wish that spraying perfume inside were equally illegal.
A male goat that sees females in heat will urinate on his beard. The smell of that is revolting outdoors and unbearable indoors, to the point that even cows (badly smelling creatures themselves) will refuse to enter a barn polluted by a perfumed goat.
A good part of human perfumes are not much more appealing to me than what a goat uses.
That's because bribes somehow have been made explicitly legal in the US.
The fix was tried in Poland in the interwar period: a law promulgated in Dz.U. 1921 nr.30 poz.177 (earlier version: Dz.U. 1920 nr.11 poz.61) said: Art.2: An official, guilty of accepting a gift or another material benefit, or a promise thereof, [a detailed list that TL;DR says "relevant to duties"], shall be punished by death by shooting.
A "campaign donation" is obviously a material benefit. No-show employment for mucho $$$s after the term ends is a promise of such benefit (and while the promise might be tricky to prove, this loophole is obvious to legislate against).
We have a president who is a full blown lunatic and he has a staff filled with criminals
You mean, warrantless wiretapping and searches haven't exploded during the previous full blown lunatic, and haven't been started by the full blown lunatic before him?
For those who lost count, those lunatics belonged to the opposite parties, and so did their staffs of criminals.
She started two wars without even being president. The Cheeto has yet to start any.
And I did not vote for her. Well, not being an USian, there's not much credit for that, but I can proudly say I did not vote for the Wicked [WB]itch. And I refuse to be fucked by a dude, especially a dog one (I type these words with a better animal on my forearms).
No, a single nuclear weapon removes the need for tens of thousands of soldiers to risk their lives. It is also drastically more cost effective. It's also the only way to reduce the loss of lives in cases like Seoul where people will have to endure "only" several hours of artillery strikes rather than weeks.
Unlike a conventional weapon which can be easily counteracted with more force, nukes reduce even an enemy who's nuclear-armed themselves from a position of "do what we tell you, or else" to an equal stance. That's why the massive investment in propaganda that tells citizens of other nations to disarm.
Actually, there's a lot of health conditions where you'll lose weight despite eating, or gain it much slower than normally. The other way, though, it's physically not possible. On uncle Adolf's diet in Auschwitz, no one gained weight no matter the "hormones". And if you use a diet more adequate than half a rotten turnip, you'll stay healthy.
And yeah, it's an addiction. Overeating is on the same boat as smoking.
Cases where BMI is badly off (ie, those with a big muscle mass) tend to be distinct from obesity. Also, more accurate methods have the downside of requiring a costly measurement, while all you need to know for BMI is weight and height.
Slashdot Mirror
Usually, the solution is: use a cracked copy (even if you have a DRM-encumbered one).
Not running adequately in Wine is in most cases self-inflicted.
One of worst offenders: google.com. Hover over a link, see where it leads. Click. Or even, left-click hold drag and cancel (esc) or right-click.
"It is estimated that one cubic meter of frozen gas hydrate contains 164 cubic meters of methane". How does 1 contain 164?
Don't be dense.
Statistically people without suitable photo ID tend to be minorities.
Except that everyone with the right to vote is in elebenty-one government databases. You had a birth certificate issued, you have a SSN, you pay taxes, in the US you can't seem to get your groceries without a car, etc. So there's more than enough ways you can point to existing government lists -- unless you're there illegally.
Which is a good part of why Democrites have their panties in a twist. Repugs have a point here.
Yeah, the breakage is not your fault. That timestamp you included might still be encoded somewhere inside the massive URL that sign-away-your-firstborn page redirects to, getting past it is currently beyond my threshold of putting up with crap.
Google somehow insists that everyone with an IP in country X prefers language X, even when accessing a country-Y-specific URL, with browser sending English as the preferred language. Searches in non-English are mostly useless except for some local stuff -- but I for one never shop/etc using Google. For plain searches there's DuckDuckGo, but there's far more to Google than that.
For a more extreme horror show, try Google over Tor. Their captchas are broken most over time.
Your link is broken -- it redirects to a country-specific YouTube site that dims the page and presents an empty white rectangle. If I recall correctly, if I dropped enough privacy-related extensions, it'd show me a form to sign away most of my rights, then keep doing so unless it's allowed setting a 3rd-party cookie. Just no.
Redirecting someone to youtube.pl when following a link to youtube.com is inexcusable, as content of these two is different (even if a good part of videos can be watched on both). They used to have an opt-out but AFAIK it's gone with no replacement.
I also have no 1h7m to waste to watch a talk I could read in 5mins or skim over in much less.
The problem is that there is not one RFC that defines an email address!! There is one that covers the top level domain (if present), one that covers the domain name (if present), and one that covers the account portion.
5322 is quite informative. The only simplification it does that I see on the first glance is that it defines "domain" as "dot-atom / domain-literal / obs-domain", dot-atom being a series of 1 or more atoms [a-z0-9!#$%&'*+/=?^_`{|}~-]+ separated by dots (ie, there might be no dots, but a dot can't be the first or last, and two dots can't come in a row). This allows using non-Internet domains, but if you know you'll use only them (a quite safe assumption), you know to refer to the DNS RFC. I'd risk saying that a policy that bans domain-literal is reasonable.
There isn't even a mandate for the @ to be present as it is not needed when sending email within one domain.
Yeah but a domain-less mail is useless on a contact form. It's about same as you said "exit our office, go left, the cafe is on the corner" vs "Sesame Street 42, Springfield TX, USA" -- valid only in a very limited number of contexts.
What makes it act like a standard is that everyone adheres to it so practically it doesn't matter if it is stuck in RFC status for eons.
e-mail RFCs are actually updated relatively often to keep being relevant, although they like to keep compat baggage.
She'll get a Lumia for Christmas.
As they're in US, that's illegal. 8th Amendment.
Depends on whether you want to support pre-Internet addresses, as that's a good part of the spec. If you drop stuff like routed addresses or comments/whitespace inside a domain, the rest is quite simple. Your MTA likely won't accept "@node.test:mary@ (this is a comment) example (whitespace around .) .net" thus it should be dropped at validation time.
Actually, most libraries are wrong, too. If you look at nodejs, it's astonishing how 90% of packages that ship a single line of code (plus tons of boilerplate) get even that single line wrong.
It doesn't take a genius to look up the relevant RFC and write a regexp.
I'll add to that that this isn't limited to web developers.
Hell yeah this. In most fields, there's decades of prior wisdom that one should at least try to get a grasp of before making "brilliant" inventions. Like, you shouldn't write an init system before reading "Unix for Dummies".
Things like SIGHUP/nohup are basic knowledge, if you don't know this you shouldn't write your snowflake way of killing processes on logout. If you don't know ways to authenticate users and what user names are explicitly allowed by POSIX (the 0day issue), you shouldn't create a gaping security hole, and even worse, you don't WONTFIX it just because one of your distribution's clicky-clicky tools doesn't allow such names, while "$EDITOR /etc/passwd" or LDAP are "user errors".
I'm for one a kernel newbie -- yet I know better than when faced with a problem like "shit network driver does high-order allocations while atomic, with disastrous reasons when under memory pressure" to rush into making an ad-hoc pool instead of reading how it should be done, or asking those with a clue. The former was my first reaction, yet I at least bothered to think before wasting developers' time with a bogus patch.
I'm quite certain passing this captcha proves you're not a human.
But the vast majority of men don't choose to become engineers either. So overall, it's more a matter of choice.
And here you nailed the most important factor.
But, when voting for a politician or hiring an employee, you don't get a time machine to their kindergarten to tell toddlers that a toy car fits girls while it's not a shame for a boy to play with dolls. No matter how hard you try to change early stages of the pipeline, any effects will be visible after more years than your company even exists. You get to pick from the pool of candidates available today.
I don't believe that's because women are any less capable
All observations show that they're indeed less capable here -- on the average. And that's not surprising: there's so many physical gender difference, such as body water content ratio, brain sizes, muscle and fat layout, ability to discern colors and smells, etc -- thus, assuming there are no mental differences would be preposterous. But, all that average tells you is that, with a gender-blind criteria, one group will make a smaller part of the population than the general count would say.
Personal variance trumps gender/race differences, thus, with a distribution without a hard cap, there will be skilled individuals from both groups above any threshold (assuming an infinitely large population).
Example of a distribution with a hard cap: D&D-like stats. Orcs get int of 3d6-2 str 3d6+2, elves 3d6+2 str 3d6-2. Thus, all the top wizards will be elves with ints 19 and 20, with not a single orc above 16. Likewise, all top fighters will be orcs. But in real life, the distribution has no such cap (ok, ok, physics tells you you can put only a certain number of processing elements and mechanical strength into a given volume, but human biology is so many orders of magnitude below physical limits that this is akin to arguing that a "no limit" stretch of Autobahn is still subject to the speed of light).
Thus: a woman picked at random is likely to be less strong than a randomly chosen man. But I still wouldn't want to get into a fight with a female lumberjack. Yet you're not going to argue that women are as capable at lumberjacking on the average, and it's only evil woodcutter companies having a bias towards men.
Oh sweet Summer Child. You actually believe Open Source is immune to politics and is a perfectly level playing field. How endearingly naive.
Well, the field obviously isn't level. Outreachy, Debconf travel funds, etc -- people indeed do get discriminated against based on their gender.
Female contributors just aren't there. I for one do quite a bit of mentoring -- out of 137 packages, there's just one upload by a woman. It was a fine upload, perfect on the first try while most people need multiple attempts. But it's the only one.
I try hard to not discriminate or even say things in a condescending tone, and I got the impression I'm doing ok -- as once the chaff is filtered out (I don't even see those who don't try to submit a package), there's no gender difference. In this case, on above corpus, women package quality is even drastically higher, but with just a single data point, this is not statistically significant.
Thus, women who do submit kernel patches or Debian packages are no worse, at all, from their male peers. The problem is, they make such a tiny portion of the submitters.
Things change once they get paid for coming. I don't believe skills of paid employees are drastically different from unpaid volunteers, thus the skew is obvious.
And what about the other tech companies, who also sport a similar overabundance? They sing in one voice here.
Which is greatly harmful to the women who actually are skilled, as they indeed tend to get dismissed by their peers as mere diversity hires. Which spins up the wheel of hate even more.
Let's compare: among 1000 kernel devs with most commits who have names that show gender (I know western and slavic naming conventions), there's 8 women. In other similar projects, gender ratios seem to be similar. Yet for example Google has 26% women among engineers. As open source projects are driven by work you actually do rather than by irrational hiring decisions, I believe it's the former figure that's representative of skill (or rather, of who decides to learn these skills). Thus, only possibilities are that 1. Google picks employees based on their race/gender rather than merit, or 2. there's a glut of skilled women somehow denied employment elsewhere. As most other big tech companies also tend to be SJW-run, they have an overabundance of diversity hires, thus we know 2. is not the case. Thus, a good part of women employed have worse skills than their peers. And thus, either they get paid less, or get unfair wage increases.
This is not to say that very highly skilled female/minority engineers don't exist -- they do, I can point you to a few. The thing is, they are rare.
so are they also banning combustion engines
Most civilized countries already announced plans to ban these by year XXXX.
perfumes, deodorants
Sadly, not yet, but see other responses.
and the millions of other products and devices that emit aerosol based chemicals.
If they're harmful or offensive to bystanders, yes, they do get banned.
This ban is for indoor. I could only wish that spraying perfume inside were equally illegal.
A male goat that sees females in heat will urinate on his beard. The smell of that is revolting outdoors and unbearable indoors, to the point that even cows (badly smelling creatures themselves) will refuse to enter a barn polluted by a perfumed goat.
A good part of human perfumes are not much more appealing to me than what a goat uses.
I am fully in favor of an Axe law.
Spraying a dangerous substance that has adverse effect (beyond just revulsion) to multiple schoolmates? Sounds pretty obvious to me.
That's because bribes somehow have been made explicitly legal in the US.
The fix was tried in Poland in the interwar period: a law promulgated in Dz.U. 1921 nr.30 poz.177 (earlier version: Dz.U. 1920 nr.11 poz.61) said: Art.2: An official, guilty of accepting a gift or another material benefit, or a promise thereof, [a detailed list that TL;DR says "relevant to duties"], shall be punished by death by shooting.
A "campaign donation" is obviously a material benefit. No-show employment for mucho $$$s after the term ends is a promise of such benefit (and while the promise might be tricky to prove, this loophole is obvious to legislate against).
Obviously, such law wasn't kept. Guess why...
We have a president who is a full blown lunatic and he has a staff filled with criminals
You mean, warrantless wiretapping and searches haven't exploded during the previous full blown lunatic, and haven't been started by the full blown lunatic before him?
For those who lost count, those lunatics belonged to the opposite parties, and so did their staffs of criminals.
She started two wars without even being president. The Cheeto has yet to start any.
And I did not vote for her. Well, not being an USian, there's not much credit for that, but I can proudly say I did not vote for the Wicked [WB]itch. And I refuse to be fucked by a dude, especially a dog one (I type these words with a better animal on my forearms).
No, a single nuclear weapon removes the need for tens of thousands of soldiers to risk their lives. It is also drastically more cost effective. It's also the only way to reduce the loss of lives in cases like Seoul where people will have to endure "only" several hours of artillery strikes rather than weeks.
Unlike a conventional weapon which can be easily counteracted with more force, nukes reduce even an enemy who's nuclear-armed themselves from a position of "do what we tell you, or else" to an equal stance. That's why the massive investment in propaganda that tells citizens of other nations to disarm.
Actually, there's a lot of health conditions where you'll lose weight despite eating, or gain it much slower than normally. The other way, though, it's physically not possible. On uncle Adolf's diet in Auschwitz, no one gained weight no matter the "hormones". And if you use a diet more adequate than half a rotten turnip, you'll stay healthy.
And yeah, it's an addiction. Overeating is on the same boat as smoking.
Cases where BMI is badly off (ie, those with a big muscle mass) tend to be distinct from obesity. Also, more accurate methods have the downside of requiring a costly measurement, while all you need to know for BMI is weight and height.