But with bugs and differences in implementation specific not only to vendors, but to models and even to firmware releases. Uniformity and standardization aren't even close to what they are for the SDK or the J2EE stack implementations.
The situation is so bad that a consortium of vendors (including sun) came up with the unified testing initiative to try to address it. On mobile platforms, you can forget about "Write once run everywhere".
I'm still going to wait for a little while before using the 4.* releases. I remember people complaining about early 3.*, and I did the same back then (waiting, not complaining). It got OK around 3.2, and i expect it to be similar with 4.*
How surprising is it that a big release takes time to stabilize ?
Exactly. It's the same thing as optimization: first code something clean and standard / best practices compliant then add the smallest amount of dirty hacks to make it fast or IE compatible.
Yes, that makes sense, but the number of such cases is infinitesimally small compared to the number of "normal" cases.
Hence the fact that the probability approaches 1. Nobody said anything about it ever reaching 1.
Not really, more about simplifying the basic stuff to be able to tackle the hard stuff.
Like using a high level programming language instead of assembly, if you want.
Why would anyone use 64-sided die when 6 coins can do the trick?
If you use 6 coins, my guess would be that you're going to toss them at the same time, so no order.
To answer the other objections, yes there are ways to go around that, my point wasn't that you can't construct an equivalent based on coins, just that simply tossing 6 coins isn't equivalent to throwing a 64-sided dice.
Speaking of overthinking...;-)
yes, i got that.
But bits are ordered, so they are distinct. Coins are not, so you have an amount of equivalent combinations in your toss, HHHTTT == TTTHHH for instance.
You can't make a difference between those, so the real case is 3 heads, 3 tails, in any order. That means that you don't have 64 cases in the end.
Well, private and protected modifiers also come to mind...
BTW:
...introduce a bug because that function was not designed or tested for use in isolation.
And in the spirit of the topic, such functions become awkward to unit test, since you're extracting a unit of work out of a loop or control structure, that logically lives there. If you face that kind of problem, your function or method is ill designed. It doesn't matter if it's a one time use or a standard API member that you are writing, a function/method should _always_ be self contained.
Separation of concern is a good practice that can (should) also be applied in non OO languages...
I know. But I still hate breathing it. I understand that, but again, in open air, it's not so much an issue.
This is about trains rather than buses: Rail industry admits that it's often greener for families to travel by car. For families perhaps, but have a look in the streets: what is the proportion of cars with only one person in, and family packed ones ?
I don't see (or smell) that in cars. What you see or smell is not the whole story. The simplest example of that is CO: no taste, no colour,... but deadly.
I'm quite sure that is not caused by CO2, but rather by ozone, and particulate pollution. CO2 concentration has an impact on breathing, it is less dangerous than CO because blood can release it, so once you stop being exposed, you start getting better. Nevertheless, high CO2 concentration can asphyxiate you without any CO presence.
CO2 levels vary very very little. Oh, the Seatle Times seems to disagree. And it's just the first relevant hit in my googling...
I'm taking the extreme position just to make a point, not because I think it will really happen that way. A negotiation tactic of sorts: you go extreme one way, I go the other, and we end up in the middle - which is where I wanted to be in the first place. I don't like that tactic. It encourages people to make less and less reasonable demands, not to try to behave and discuss like adults. That's a bit like the pre-emptive strike tactic.
It's not true what you say about cars. If you use sulfur-free fuel there is technology to make the exhaust cleaner then what went in. The exhaust is just CO2 and H2O, no NOx's, no soot or unburnt fuel. There's always the car in itself once its life cycle terminates, which pollutes quite a lot and that pollution has a non zero impact on your health.
And you quite missed the point with cholesterol and sport.
I am not trying to make anyone else healthy, it's none of my business. I just don't want them to send smoke to me. I couldn't care less if they smoke - just keep the smoke out of the air I breathe. And that includes when I walk on the sidewalk. Point taken. My opinion is that smokers in open air are too small a health problem to impose such a harsh restriction. Especially compared to the impact of other factors, such as transport and general consumption-related pollution (plastic packages come to mind, they get burned after being collected which creates far more toxic fumes than cigarettes.)
(BTW just to clarify: my almost 0% number was not amount of pollution, it was amount of health effects produced.) Moot point, they are closely related.
Also, why do you say 'individual' cars? Buses are worse as far as pollution goes. Are you saying a single individual in his car pollutes less than bus passengers per capita ? where are the numbers, I'd like to see them...
And just so you know that I am consistent, I would also like to ban coal power plants (due to mercury and radioactivity released), and replace them with nuclear.
I once had ideas of an air pollution tax: every item sold, from hairspray and bleach to paint and kerosene, and including electricity (based on the individual power plant - the tax is collected at point of production, not consumption, but it would be passed on), would carry a very visible tax. The amount of the tax would be calculated based on expected health care costs of the estimated emissions of the product. The money collected would go to all residents, citizen or not, divided equally with no regard to income.
So if everyone used about the same amount of polluting products the net effect would be 0. But in actuality the worst emitters would pay more, and market forces would quickly remove the worst pollutants from the market. Nice ideas, that are slowly being put into practice (Kyoto, some other national-level laws...). It remains to be seen how efficient that will/would be. But I agree on principle, and I think there's a much much bigger public health gain at stakes there.
I would NOT include CO2 in the calculations. Only emissions with health effects. I think that's a mistake, CO2 has health impacts, especially when it's kept down by smog-like conditions. Ask an asthmatic how he feels about it...
It's nice to be consistent, but be careful about extremes like "0% or nothing". Usually the best solution lies in between the extremes, rarely at them.
My whole point is that you won't be able to reduce car pollution to "nothing at all", just the same as you can't smoke anything without "letting any of it into the air". There's always a residue, and your "any number over 0 is unacceptable" fails to take that into account.
BTW: I think it would be closer to reality to say "There is technology that is making gas using cars cleaner" than "very clean".
As for entirely banning smoking, OK, but then you also ban individual cars, alcohol and "cholesterol filled" foods, amongst others, otherwise it's just hypocrisy. And while we're at it we could also make a minimum amount of sport mandatory.
The example was obviously badly expressed. Do you deny that cars cause health problems ? Smog, etc are related to lung sicknesses too.
But anyway, you missed my point.
I too enjoy going to the restaurant better without smoke, but this:
but inhaling particles of carbon in the lungs is not good for anyone.
It doesn't matter is the stats say 0.01% of people are harmed - those people did not have a choice, so any number over 0 is unacceptable. in your original post is what got me started. The so any number over 0 is unacceptable part is a bit extremist, and makes it look like you don't understand what "error margin" means. The inhaling particles of carbon in the lungs is not good for anyone also clearly lacks of discrimination between lots of CO-CO2 producing activities in wich i am certain you take part (hence the car example).
Do you drive a car ? And dou you breathe ? Because i don't know if you're aware, but breathing releases CO2 too, you know...
I don't care what you believe, but any number over 0 is unacceptable, so please stop breathing around me.
Seriously, though, I am a smoker and all in favour of non-smoking regulations in restaurants. I do enjoy a smokeless meal (unless I'm having salmon:P ) but I can't stand the witch-hunt currently going on against smokers.
Slashdot Mirror
But with bugs and differences in implementation specific not only to vendors, but to models and even to firmware releases.
Uniformity and standardization aren't even close to what they are for the SDK or the J2EE stack implementations.
The situation is so bad that a consortium of vendors (including sun) came up with the unified testing initiative to try to address it.
On mobile platforms, you can forget about "Write once run everywhere".
Well, the UK has fought successfully in the south before. I wouldn't treat their claim too lightly.
I think you're referring to The Last Starfighter
Kudos on the mess equality thing :-D
I'm still going to wait for a little while before using the 4.* releases. I remember people complaining about early 3.*, and I did the same back then (waiting, not complaining). It got OK around 3.2, and i expect it to be similar with 4.*
How surprising is it that a big release takes time to stabilize ?
Exactly. It's the same thing as optimization: first code something clean and standard / best practices compliant then add the smallest amount of dirty hacks to make it fast or IE compatible.
Set the plain old text mode in the options. ;)
No idea how to do syntax highlighting, though
Yes, that makes sense, but the number of such cases is infinitesimally small compared to the number of "normal" cases.
Hence the fact that the probability approaches 1. Nobody said anything about it ever reaching 1.
Undoing (hopefully) a mis-moderation. Wanted to moderate insightful, ended up clicking redundant.
Sorry
Not really, more about simplifying the basic stuff to be able to tackle the hard stuff.
Like using a high level programming language instead of assembly, if you want.
I only have Euros, you insensitive clod :P
Why would anyone use 64-sided die when 6 coins can do the trick?
If you use 6 coins, my guess would be that you're going to toss them at the same time, so no order. ;-)
To answer the other objections, yes there are ways to go around that, my point wasn't that you can't construct an equivalent based on coins, just that simply tossing 6 coins isn't equivalent to throwing a 64-sided dice.
Speaking of overthinking...
yes, i got that.
But bits are ordered, so they are distinct. Coins are not, so you have an amount of equivalent combinations in your toss, HHHTTT == TTTHHH for instance.
You can't make a difference between those, so the real case is 3 heads, 3 tails, in any order. That means that you don't have 64 cases in the end.
I see the funny side of your comment, but 6 coins aren't equivalent to a 64 dice: they are indistinguishable so HHHTTT == TTTHHH.
BTW:
...introduce a bug because that function was not designed or tested for use in isolation.And in the spirit of the topic, such functions become awkward to unit test, since you're extracting a unit of work out of a loop or control structure, that logically lives there. If you face that kind of problem, your function or method is ill designed. It doesn't matter if it's a one time use or a standard API member that you are writing, a function/method should _always_ be self contained.
Separation of concern is a good practice that can (should) also be applied in non OO languages...
And you quite missed the point with cholesterol and sport.
I am not trying to make anyone else healthy, it's none of my business. I just don't want them to send smoke to me. I couldn't care less if they smoke - just keep the smoke out of the air I breathe. And that includes when I walk on the sidewalk. Point taken. My opinion is that smokers in open air are too small a health problem to impose such a harsh restriction. Especially compared to the impact of other factors, such as transport and general consumption-related pollution (plastic packages come to mind, they get burned after being collected which creates far more toxic fumes than cigarettes.) (BTW just to clarify: my almost 0% number was not amount of pollution, it was amount of health effects produced.) Moot point, they are closely related. Also, why do you say 'individual' cars? Buses are worse as far as pollution goes. Are you saying a single individual in his car pollutes less than bus passengers per capita ? where are the numbers, I'd like to see them... And just so you know that I am consistent, I would also like to ban coal power plants (due to mercury and radioactivity released), and replace them with nuclear.
I once had ideas of an air pollution tax: every item sold, from hairspray and bleach to paint and kerosene, and including electricity (based on the individual power plant - the tax is collected at point of production, not consumption, but it would be passed on), would carry a very visible tax. The amount of the tax would be calculated based on expected health care costs of the estimated emissions of the product. The money collected would go to all residents, citizen or not, divided equally with no regard to income.
So if everyone used about the same amount of polluting products the net effect would be 0. But in actuality the worst emitters would pay more, and market forces would quickly remove the worst pollutants from the market. Nice ideas, that are slowly being put into practice (Kyoto, some other national-level laws...). It remains to be seen how efficient that will/would be. But I agree on principle, and I think there's a much much bigger public health gain at stakes there. I would NOT include CO2 in the calculations. Only emissions with health effects. I think that's a mistake, CO2 has health impacts, especially when it's kept down by smog-like conditions. Ask an asthmatic how he feels about it...
It's nice to be consistent, but be careful about extremes like "0% or nothing". Usually the best solution lies in between the extremes, rarely at them.
My whole point is that you won't be able to reduce car pollution to "nothing at all", just the same as you can't smoke anything without "letting any of it into the air". There's always a residue, and your "any number over 0 is unacceptable" fails to take that into account.
BTW: I think it would be closer to reality to say "There is technology that is making gas using cars cleaner" than "very clean".
As for entirely banning smoking, OK, but then you also ban individual cars, alcohol and "cholesterol filled" foods, amongst others, otherwise it's just hypocrisy. And while we're at it we could also make a minimum amount of sport mandatory.
Do you drive a car ? And dou you breathe ? Because i don't know if you're aware, but breathing releases CO2 too, you know... I don't care what you believe, but any number over 0 is unacceptable, so please stop breathing around me. :P ) but I can't stand the witch-hunt currently going on against smokers.
Seriously, though, I am a smoker and all in favour of non-smoking regulations in restaurants. I do enjoy a smokeless meal (unless I'm having salmon