Ripper
Зичу Вам успіхів
- Joined
- November 8, 2014
- Messages
- 12,085
Baby number 2 has arrived. Another son
So no sleep for me for the rest of my life, but he's such a cute little potato head that it's ok.
Mazel tov!
- Joined
- Nov 8, 2014
- Messages
- 12,085
Mazel tov!
Baby number 2 has arrived. Another son
So no sleep for me for the rest of my life, but he's such a cute little potato head that it's ok.
A couchpotato head?
pibbur who congratulates the pladio and remembers that both (I think there are two of them) his kids (girls) has passed 40.
It wasn't me I swear. I'm not the father. I can't be.:lol:
Joking aside Congratulations Pladio.
The Who - It's a Boy
Joking aside Congratulations Pladio.
The Who - It's a Boy
Baby number 2 has arrived. Another son
So no sleep for me for the rest of my life, but he's such a cute little potato head that it's ok.
Congrats!
Congratulations !
Pladio, you sneaky devil... Congrats!
Yesterday it was exactly 67 years since the first Fortran program saw the light of day, on the 20th of september 1954 (about 3 months before I saw the light of day). See https://www.edn.com/1st-fortran-program-runs-september-20-1954/.
Fortran, developed by a team at IBM lead by John Backus, is considered the first high level programming language. Regarding his motives Backus said in 1979: ""Much of my work has come from being lazy. I didn't like writing programs, and so, when I was working on the IBM 701, writing programs for computing missile trajectories, I started work on a programming system to make it easier to write programs" (Wikipedia).
Initially, people were a bit reluctant, fearing that the language would perform worse than code written directly in assembly. However, that change when the first commercially avaliable compiler, an optimizing compiler, was released in 1957, and people realized that one line of Fortran could on the averagde replace 20 assembly statements.
Fortran is still in use, especially for scientific programming. In August 2021 it was ranked 13 in the TIOBE list of the most popular programming languages. I suspect this is largely due to nostalgie, since it's position a year before was 42. I learned Fortran dring my computer science days at university in the early '90s. I haven't used it since, although, AFAIR, I did translate some Fortran code into C back then.
I've tried to find what the program did, but so far I haven't found it.
pibbuR who has a Fortran formula (Euler's identity) written on his guitar.
Fortran, developed by a team at IBM lead by John Backus, is considered the first high level programming language. Regarding his motives Backus said in 1979: ""Much of my work has come from being lazy. I didn't like writing programs, and so, when I was working on the IBM 701, writing programs for computing missile trajectories, I started work on a programming system to make it easier to write programs" (Wikipedia).
Initially, people were a bit reluctant, fearing that the language would perform worse than code written directly in assembly. However, that change when the first commercially avaliable compiler, an optimizing compiler, was released in 1957, and people realized that one line of Fortran could on the averagde replace 20 assembly statements.
Fortran is still in use, especially for scientific programming. In August 2021 it was ranked 13 in the TIOBE list of the most popular programming languages. I suspect this is largely due to nostalgie, since it's position a year before was 42. I learned Fortran dring my computer science days at university in the early '90s. I haven't used it since, although, AFAIR, I did translate some Fortran code into C back then.
I've tried to find what the program did, but so far I haven't found it.
pibbuR who has a Fortran formula (Euler's identity) written on his guitar.
Baby number 2 has arrived. Another son
So no sleep for me for the rest of my life, but he's such a cute little potato head that it's ok.
Congrats!
Huh! I’ve been paid to convert programs in Fortran 2 to Fortran 4 ( late 1977). Interesting language.
I personally think Backus’ major contribution to computing was BNF, but I’m biased I though Algol was amazing, still more advanced than most available languages now.
Thanks all
tl:dr: I promised I wouldn't say more about statiscs. But that was in the lackblogger thread. Here it's another matter. LOTS OF WORDS coming up.
First let me define a few terms (which you admittedly may already know). A population is all the people (or dogs or things or…) that we want to know somethng about. For instance the the number people buying a game, and we want to know how much they liked it. For a gaame that sold millions of copies it's a bit unpractical to ask everyone, so in stead we examine a much smaller group, hoping that it will tell us what we would like to know. About the population. We call that group a sample, and we may get the sample by asking a number of people, or we may pick a ready-made sample, like those reviewing/scoring a game on Metacritic. To learn something reliable about the population, the sample has to be representative, and for that there are at least two important requirements. The sample must be large enough and it must satisfy the demand for statistical independence. It not we say that the sample is biased.
I mentioned that when it comes to user reviews, statistical independence is about how opinions affect how likely it is that someone publish a review/score. And I said that it really would surpise me if those two factors aren't connected. Someone with strong opinions are more likely to post than those just satisfied with a game. This will affect how representative the sample is. How much bias there is may be difficult to estimate, but the sample of user reviews will be (to some degree) biased. Which makes knowing how people in general think difficult.
Another example. On one of our TV stations, when there was a debate in the studio, viewers were invited to phone in/send SMS's "voting" on the questions debated. Once there were around 30 000 callers, and the host said that this had to be more representative for public opinion, compared to the much smaller samples polling agencies worked with. Sadly no. Again, the likelyhood of people calling will most likely be strongly affected by their opinions. And a huge biased sample is still a biased sample, and many statistical analysis methods collapse. So what we can know for sure in this case is the opinion of the callers, but to a (much) smaller degree what the Norwegian people think.
So, what about polling agencies? Well, you don't call polling agencies. They call you, so your opinion has less concequences for giving your opinion. There's still the matter of whether your opinions affect your willingness to answer, but I assume that factor is smaller, and they may take that into account in their models. By carefully designing the set of criteria for including people, they can operate with much smaller samples and stil get fairly reliable results (over time). They don't always succeed. AFAIK most pollers wrongly predicted that Hilary Clinton would win the president election. One explanation could be that they didn't catch enough of the typical Trump voters in their sample. Or that more Trump-people than expected voted. NB! These are just my (hopefully educated) guesses.
Now for something completely different: Asking the correct or the wrong questions. In Norway there is a diet supply, let's call it VP, which the vendor claims is very important for our well-being. I think it's humbug, but that's not important. This is: One argument they have for why it works is that 87% of those using it are happy with it. Which makes me wonder about those 13%: Why are they using a diet supply they're not happy with? (It's a bit like: I hate Skyrim (with or without mods), but I still play it after all these years). It says nothing. What would be interesting is how many have bought VP and stopped using it compared to the number who keep using it. We don't get those numbers.
Another example. Back in '99 I was part of a project introducing a PACS system for my X-ray department. PACS (Picture Archive and Communication System) is an electronic x-ray system, where every image is stored digitally, and easy obtainable. One part of the project was evaluating the result of introducint PACS at the hospital, and the one responsible designed a test where users were asked to evaluate how happy they were on a scale from 1 (very bad) to 5(very good) before and after PACS. It was useless (and I told them, unsuccessfully). The average score before PACS was 4/5. How could you then detect an improvement? There are several factors where a PACS system objectively (if I may use that word) and significantly is an improvement. For instance: Every day clinicians come to the X-ray department, and a radiologist presents the images taken for their (clinician's) department during the last 24 hours. Before PACS the original x-rays films were shown on - don't know the English word - frames with backlighting. Those sitting at the back of the room would have real problems seeing those images. With PACS, images are enlarged and shown on a projection screen, so that everyone can see. Another issue: Quite often the clinicians want to see images not selected for the presentation. In the old days they then had to wait until the next day (sometimes longer), because those images were in one of the physical archives, and not readily available. With PACS it's just a quick search, and the images are available in a couple of seconds. So, introducing PACS clearly improves the service for the clinicians, but with a pre PACS satisfaction score of 4/5, there's no way that improvement could be detected. I wouldn't be surprised if the post PACS score would be about the same, because the clinicians wouldn't use the same scale after PACS. (When answering queries like this, I for instance tend to avoid the extreme happines/unhappiness scores). More correct questions would be: How often have you trouble seeing the images presented? And how often do you ask for other examinations and how long do you have to wait to see them?
One final example. One of our real estate agencies (they're out of busines now) presented the result of this public enquiry: Which agency is know for getting the highest price for the seller? And surprise, surprise it was them. Now first of all, that an agency is in the eye the public is "known" to achieve high prices does not mean that they actually achieve higher prices. Secondly, how would the public know? The price achieved depends on a lot of factors independent of the agency: Timing, location (location, location) and the condition of the property. How would people know how much those factors affected the price? So what the company really asked is "what agency is known for claiming that they achieve a better price. Which is not very useful.
There are other factors I'd like to touch upon, but I'll leave that for tomorrow. This post have been more than long enough.
Puh!
pibbuR who is the sole sample and the sole population of pibburs.
PS. One observation: If it's true that the agency achieved higher prices, as a buyer you should clearly stay away from properties sold by them, since you will get better prices buying elsewhere. DS.
First let me define a few terms (which you admittedly may already know). A population is all the people (or dogs or things or…) that we want to know somethng about. For instance the the number people buying a game, and we want to know how much they liked it. For a gaame that sold millions of copies it's a bit unpractical to ask everyone, so in stead we examine a much smaller group, hoping that it will tell us what we would like to know. About the population. We call that group a sample, and we may get the sample by asking a number of people, or we may pick a ready-made sample, like those reviewing/scoring a game on Metacritic. To learn something reliable about the population, the sample has to be representative, and for that there are at least two important requirements. The sample must be large enough and it must satisfy the demand for statistical independence. It not we say that the sample is biased.
I mentioned that when it comes to user reviews, statistical independence is about how opinions affect how likely it is that someone publish a review/score. And I said that it really would surpise me if those two factors aren't connected. Someone with strong opinions are more likely to post than those just satisfied with a game. This will affect how representative the sample is. How much bias there is may be difficult to estimate, but the sample of user reviews will be (to some degree) biased. Which makes knowing how people in general think difficult.
Another example. On one of our TV stations, when there was a debate in the studio, viewers were invited to phone in/send SMS's "voting" on the questions debated. Once there were around 30 000 callers, and the host said that this had to be more representative for public opinion, compared to the much smaller samples polling agencies worked with. Sadly no. Again, the likelyhood of people calling will most likely be strongly affected by their opinions. And a huge biased sample is still a biased sample, and many statistical analysis methods collapse. So what we can know for sure in this case is the opinion of the callers, but to a (much) smaller degree what the Norwegian people think.
So, what about polling agencies? Well, you don't call polling agencies. They call you, so your opinion has less concequences for giving your opinion. There's still the matter of whether your opinions affect your willingness to answer, but I assume that factor is smaller, and they may take that into account in their models. By carefully designing the set of criteria for including people, they can operate with much smaller samples and stil get fairly reliable results (over time). They don't always succeed. AFAIK most pollers wrongly predicted that Hilary Clinton would win the president election. One explanation could be that they didn't catch enough of the typical Trump voters in their sample. Or that more Trump-people than expected voted. NB! These are just my (hopefully educated) guesses.
Now for something completely different: Asking the correct or the wrong questions. In Norway there is a diet supply, let's call it VP, which the vendor claims is very important for our well-being. I think it's humbug, but that's not important. This is: One argument they have for why it works is that 87% of those using it are happy with it. Which makes me wonder about those 13%: Why are they using a diet supply they're not happy with? (It's a bit like: I hate Skyrim (with or without mods), but I still play it after all these years). It says nothing. What would be interesting is how many have bought VP and stopped using it compared to the number who keep using it. We don't get those numbers.
Another example. Back in '99 I was part of a project introducing a PACS system for my X-ray department. PACS (Picture Archive and Communication System) is an electronic x-ray system, where every image is stored digitally, and easy obtainable. One part of the project was evaluating the result of introducint PACS at the hospital, and the one responsible designed a test where users were asked to evaluate how happy they were on a scale from 1 (very bad) to 5(very good) before and after PACS. It was useless (and I told them, unsuccessfully). The average score before PACS was 4/5. How could you then detect an improvement? There are several factors where a PACS system objectively (if I may use that word) and significantly is an improvement. For instance: Every day clinicians come to the X-ray department, and a radiologist presents the images taken for their (clinician's) department during the last 24 hours. Before PACS the original x-rays films were shown on - don't know the English word - frames with backlighting. Those sitting at the back of the room would have real problems seeing those images. With PACS, images are enlarged and shown on a projection screen, so that everyone can see. Another issue: Quite often the clinicians want to see images not selected for the presentation. In the old days they then had to wait until the next day (sometimes longer), because those images were in one of the physical archives, and not readily available. With PACS it's just a quick search, and the images are available in a couple of seconds. So, introducing PACS clearly improves the service for the clinicians, but with a pre PACS satisfaction score of 4/5, there's no way that improvement could be detected. I wouldn't be surprised if the post PACS score would be about the same, because the clinicians wouldn't use the same scale after PACS. (When answering queries like this, I for instance tend to avoid the extreme happines/unhappiness scores). More correct questions would be: How often have you trouble seeing the images presented? And how often do you ask for other examinations and how long do you have to wait to see them?
One final example. One of our real estate agencies (they're out of busines now) presented the result of this public enquiry: Which agency is know for getting the highest price for the seller? And surprise, surprise it was them. Now first of all, that an agency is in the eye the public is "known" to achieve high prices does not mean that they actually achieve higher prices. Secondly, how would the public know? The price achieved depends on a lot of factors independent of the agency: Timing, location (location, location) and the condition of the property. How would people know how much those factors affected the price? So what the company really asked is "what agency is known for claiming that they achieve a better price. Which is not very useful.
There are other factors I'd like to touch upon, but I'll leave that for tomorrow. This post have been more than long enough.
Puh!
pibbuR who is the sole sample and the sole population of pibburs.
PS. One observation: If it's true that the agency achieved higher prices, as a buyer you should clearly stay away from properties sold by them, since you will get better prices buying elsewhere. DS.
Last edited:
I (I think) have previously recommended the site https://www.worldsciencefestival.com which presents a lot of interesting science stuff, mainly in the form of discussions (talkshows) with well known scientists (often lead by Brian Greene).
Recently I watched this show: "The Matter Of Antimatter: Answering The Cosmic Riddle Of Existence", which included a rather thorough discussion of the so called "baryon assymetry" problem. The universe seems to be made of matter. When antimatter is produced today, there's always symmetry. We get one matter particle and one antimatter particle. And matter and antimatter annihilate when they quickly collide. Therefore there really shouldn't be a universe, if at the time of the big-bang (or however the universe came to be), matter and antimatter were produced in equal quantity. So there had to be an assymetry back then, there were produced more matter than antimatter. This is one of the great paradoxes in physics, and nobody knows how this came to be. Very interesting.
Another interesting thing which came up: The main production facility of antiprotons today is the huuuge particle accelerator in (at?) CERN, and their production may seem to be significant, they produce on the average around 100000000000000 (10^14) antiprotons each year. Unfortunately 1 gram of antiprotons contains around 6*10^23 protons (Avogadro's number). Which means at the current production rate it takes around a billion years to make 1 gram of the stuff. And it would be hideously expensive. So unfortunately, while it would be very efficient, there are still a few practical problems using antimatter as fuel.
pibbuR who, despite his somewhat negative outlook on life doesn't destroy (at least only slightly) positive people.
Recently I watched this show: "The Matter Of Antimatter: Answering The Cosmic Riddle Of Existence", which included a rather thorough discussion of the so called "baryon assymetry" problem. The universe seems to be made of matter. When antimatter is produced today, there's always symmetry. We get one matter particle and one antimatter particle. And matter and antimatter annihilate when they quickly collide. Therefore there really shouldn't be a universe, if at the time of the big-bang (or however the universe came to be), matter and antimatter were produced in equal quantity. So there had to be an assymetry back then, there were produced more matter than antimatter. This is one of the great paradoxes in physics, and nobody knows how this came to be. Very interesting.
Another interesting thing which came up: The main production facility of antiprotons today is the huuuge particle accelerator in (at?) CERN, and their production may seem to be significant, they produce on the average around 100000000000000 (10^14) antiprotons each year. Unfortunately 1 gram of antiprotons contains around 6*10^23 protons (Avogadro's number). Which means at the current production rate it takes around a billion years to make 1 gram of the stuff. And it would be hideously expensive. So unfortunately, while it would be very efficient, there are still a few practical problems using antimatter as fuel.
pibbuR who, despite his somewhat negative outlook on life doesn't destroy (at least only slightly) positive people.
When at was at the university, they were still teaching Fortran because of the amount of existing libraries. Apparently many scientists were/are still using it, although hopefully not in their '77 or earlier flavours.
As everywhere I guess, we were told about the famous Fortran bug that lost a spacecraft, Mariner I, because of a typo. I recently read that it was apparently just a hoax.
It could have happened, but the error was detected. Code validation at its best.
One thing I noticed I'm not even an official news-poster on on RPGWatch anymore, and I still get shit comments on my personal news-thread. So your not alone @Redglyph;.
Just a random thought today because a few members bitched about watching video news. Anyway this is not important just blowing off some steam. So pleae ignore me.
It fits the things you don't need to know thread.
Just a random thought today because a few members bitched about watching video news. Anyway this is not important just blowing off some steam. So pleae ignore me.
It fits the things you don't need to know thread.
Yeah I saw that...One thing I noticed I'm not even an official news-poster on on RPGWatch anymore, and I still get shit comments on my personal news-thread. So your not alone @Redglyph;.
Just a random thought today because a few members bitched about watching video news. Anyway this is not important just blowing off some steam. So pleae ignore me.
It fits the things you don't need to know thread.
But it was quite decent though, wasn't it?If you want an indecent newsbit, post the last video from NeverKnowsBest. It will surely generate some traffic wherever it's posted :lol: I'm usually posting his videos, even if they're stretched a bit too long for the subject sometimes. Maybe not this time (I know, that last sentence is ambiguous, it's on purpose).
So true at least my were decent enough not to personally attack me. Anyway I enjoy those videos you post, and you explained abut how they define CRPG differently.
Anyway it amazes me how many can't be bothered to watch a 2-3 minute video.:lol:
Anyway it amazes me how many can't be bothered to watch a 2-3 minute video.:lol:
lackblogger
SasqWatch
- Joined
- November 1, 2014
- Messages
- 4,778
According to official statistics, in the UK at least, fewer people topped themselves during lockdown as compared to previous years.
This might suggest that more people get traumatised by chronic socialising than chronic loneliness.
Another stab in the heart for those with the gregarious trait
This might suggest that more people get traumatised by chronic socialising than chronic loneliness.
Another stab in the heart for those with the gregarious trait
- Joined
- Nov 1, 2014
- Messages
- 4,778