I've had this discussion with my girlfriend many times before: when we want to stream HD content across the wireless network reliably from one room to another to support our video on demand system, we'll have to upgrade our wireless router.
Oh, you mean people have a need for bandwidth on the LAN, even though they're only connected to a 10Mb internet connection?
I've had reviewers give me guff while reviewing population genetics papers because the reviewer didn't understand the basics of statistics. That's a field I shouldn't have to explain. If you can't grok a Chi-square test, then you should get out of the field, or at least not review papers.
You have to write for your audience, and assume some level of knowledge. You also often have to deal with word limits, so you can't write as much as you'd want to.
I have a friend who worked in the hedge fund game for a number of years. He's a brilliant mathematician, and worked on the models they used to inform their trading. The group he worked with was quite successful, and make a heck of a lot of money.
One of his most interesting comments: "The model can inform your decisions, but you have to know when to NOT trust the model." Another of his comments on a completely different talk: "Mathematical models are never perfect, but they can be useful."
The trading system can be modeled, but you can never capture all the true complexity of the real world. If you leave the model to do it's thing, if I know how it's going to act, I can game the system. If the world changes in a way that the model builders did not predict, then the system will also act inappropriately.
I can't imagine ever getting rid of all the traders out there, though I imagine expert systems will become more 'expert' as time goes on.
When you say there's nothing to parody by bad acting, you're referring to the movie Scream (1-4 is it now?), I assume. Or any other play on B movies (or you could parody bad writing with a novel like "Snow crash", which was just a parody of all the crappy cyperpunk before it.) My point is, "bad" gives plenty to parody. Generally with "much worse", but not with the seriousness.
(as an aside, I think people who create a bad but don't take themselves seriously are hard to parody, while it's people who think they're king of all they survey and turn out crap are easy targets...)
What Colbert does, is take on all of the bullshit attitude of some of the people you mention (especially O'Reilly), and say things that point out how rediculous the attitude is. Some examples from wikipedia:
The Colbert Report directly parodies The O'Reilly Factor with a commentary segment called "The Wørd", similar to O'Reilly's "Talking Points Memo". Like "the Memo", "The Wørd" features Colbert asserting a political point of view with a text screen graphic next to him. However, while O'Reilly's text serves to emphasize his points, Colbert's generally serve to provide an ironic counterpoint to his position. Also, Colbert's "No Fact Zone" is clearly inspired by O'Reilly's "No Spin Zone".
-------
Or, take for example the word "truthiness":
------- In "The WØRD" on the first episode of the Report, Colbert featured the term "truthiness", which he defined as "The quality by which one purports to know something emotionally or instinctively, without regard to evidence or intellectual examination"
-------
If you don't find the attitude and over-the-top distortion amusing, then perhaps you just don't like the satire. I understand many people don't get satire (there's an interesting scientific study about satire and who does and doesn't get it that was published a few years ago), or perhaps it's just not your style. In either event, that's acceptable, and the above comments are not a flame of your preferences.
"So it comes to this: you cheer on Colbert because he "skewers" people with whom you disagree. "
You should watch O'Reilly, so you can see that Colbert is a paradody of the right wing "I'll shout louder than you" argumentation style. Heck, listen to talk radio for a little while, and you'll get the idea.
It's a parody of a style of 'conversation', where there's no dialog taking place, just spewing of the party line, louder than the other person. Besides, if you don't like what the other person is saying, you just shut their mike off.
Strangely (?) I read that you can fix this coil whine with a little superglue. Someone claimed that you could immobilize the coils so they didn't move, and it would not effect their operation.
Unfortunately, I don't rememeber if I read that here, or on the unofficial thinkpad forum, or somewhere else, so I don't have a reference. Perhaps someone with better hardware knowledge than I could confirm or deny this rumor.
My experience, while limited, is that if you pick the right company, then they will honor their warantee and fix the computer while it's in warantee. The question is: how long do you want to have a warantee for?
Dell fixed our laptops for 3 years. The laptops had many more problems than I would have liked (probably 12-15 repairs on 5 laptops over 3 years.) Now that the warantee on them is over, we'll replace them as they die with something that is in warantee.
The new thinkpads we're buying all have 3 year complete care warantees. I dropped a glass of water in one, and it's being fixed for free. I expect we'll get 3 years out of those laptops, and any extra time is a bonus.
I think the question of warantees that is important is: How reasonable are the service people, and how much work do you have to do to get your computer serviced? Dell makes you jump through hoops, and they seem to screw up our server more than they fix it. Lenovo seems to take you at your word, and fix things ASAP. I must admit that I also like the fact that lenovo's phone support has enough people staffing it that you get through quickly, and the call centers are in the country I'm in (the US), so I have no communications gap. Dell seems to do most of their support over IRC-style chat, or email, and the gaps in communication are much wider.
Holy Crap. I lived in Cary, NC for a year. I couldn't get out quick enough. Sure, good salary, and good housing. No nightlife, no good food, you had to drive 10-20 miles to get anywhere. I'm originally from NY, and that's not my idea of a 'city'. Boston suits me much better, and while 'biotech' is big in RTP, the largest research centers are in Boston/San Fan. RTP seems to have as much manufacturing as research - which makes it sound like a 'center', but there are not nearly as many PhD's in the area as you'd think.
I'd have to wonder if any of the other areas are any better.
Plus, I'd hate to have to mail order all my food. I love to eat, and large cities tend to have resturaunts and grocery stores for immigrant populations. I love to go to the local brazillian, or japapese, or korean, or pakistani grocery store to buy all those ingredients you can't get at a chain store.
I think a lot of smaller cities offer good ratio of salary to housing, but what if you need more?
Actually, in many cases, the 'pure luck' option now is actually 'test all possible causes of type '. For example, if I'm going to study Alhemiers with the intent of finding the genetic variation responsible for the disease, I don't want to read literature papers, pick genes *I* think are involved, and just look at those areas. The emerging trend now is to test 95%+ of all common variation in the human genetic sequence, and find the variation that tracks with the disease. You then work *back* to functional information about how those variations affect the biology.
I'd say "understanding how the brain works" is, for some kinds of science, the old school method. Why? Because our understanding is incomplete, our tests using best guesses haven't yielded a lot of luck, so why not explore the whole problem space? Actually, as one of my well known colleges says "If there's one thing we've found out over the last 10 years, it's that we're exceptionally bad at picking regions to study."
I think putting up graphs without even the kindness of a legend, labels, etc is not terribly useful. And while I'm not in the protein folding field, I'm close by in genetics, and we look at protein folding and how mutations affect active sites.
You need to re-read this bit:
but if they do not convey information (and not lose a large amount of relevant information) then they are just a nice way to generate patterns for some nerd's tie.
Like I said in another post, these guys ought to read a bit of Tufte, so the graphs are usable. Unless of course this article blatently cuts and pastes only select parts of the graph that look "pretty". In which case, the authors of the article are the people who have rendered them useless.
Thanks for letting me know what I don't do for a living, though, as I'd nearly forgotten and thought I was a professional chef.
When I read articles about biology, especially molecular biology/genetics, I see lots of interesting "facts" about the field given by various members of the slashdot crowd. I'm not a leader in the field, but I feel knowledgeable enough working in the field to know just how wrong these "facts" are, yet get modded insightful.
What scares me are all the articles about topics that I'm not an expert in, where I can't judge the veracity of comments. I've realized that if you guys are so terribly wrong here, that you're probably not believeable anywhere else, either.
Not that this news to anyone. It just depresses me everytime I see this type of story come up.
I don't buy junk DNA. The reason? There's lots of regions that are not in coding regions, but are conserved (read: the same) with other organisms much higher than chance would predict. These regions have been tested, and found that they are being kept the same under via selection.
Want to know more? Read Nat Genet. 2006 Feb;38(2):223-7.
In the interest of full disclosure: I'm one of the authors, and I'm sick of reading about 'junk dna'.
The thing is, if it's very hard to communicate the information without training to interpret it, then you'd probably want to go back and refactor the design to increause the usability. Does no one read Tufte anymore?
I haven't had to cast an object since I started using java 1.5 18 months ago. Generics, look into them. Also, you can put primitives into collections (via autoboxing).
Java 1.5 made the language a LOT nicer and more expressive.
This is exactly right. This is why there's a methods/materials section in papers, so that we have the reagents and the process neccesary to reproduce the results you put in the discussion section.
The point of visualizing data is to learn something that you could not do with the raw data. In all of the cases shown in the article (yes, I acually read TFA), I didn't spot an example where it actually showed anything useful.
The first example with proteins: how similar are two proteins? If two shapes are similar (and please, how many proteins where being graphed there? One, two, five?), then you might be able to recognize it. If they are similar shapes, are they always presented in the same orientation in space? Does color have any meaning? Does this graph have any legend? If I gave someone who understood the graphs two proteins, what could he say besides "these are related" and "these are not related"? We already have wonderful programs to compare two proteins and say how similar they are two each other, along with being able to the estimate significance of the measurement.
I'm not sure that the other graphics look more informative. They are all pretty, but if they do not convey information (and not lose a large amount of relevant information), then they are just a nice way to generate patterns for some nerd's tie.
I put a lot of stock in the fact that some politicians can talk, and express their ideas, instead of just releasing press kits.
Listen to bush some time, and then read his press kits. While he's trying desperately to repeat what's in the press kits (until he goes 'off script' and makes a fool of himself), other officials have a lot more elegance and charisma. Listen to Barack OBama speak - that man has passion, intensity, and intelligence. That's the sort of guy I want out in the world, trying to improve our foreign policy.
The Broad institute has 454 as one of their sequencing platforms in house. Of course, george church is working on an 'open source' sequencing platform that will be obscenely cheap by comparison. I think you need a microscope, some (comparably) reagents, and some open source software to do sequencing.
You might remember the Broad institute when they were the whitehead center for genome research - the people who sequenced a large (the largest?) part of the human genome. A number of people there seem to think that the 454 technology is at least 'pretty interesting'.
I'm interesting in anyone who can do the $1000/person sequencing. We're using Affy 500K chips now to do whole genome scans, but why not get the full sequence (and find those out of LD, really rare SNPs as well as indels...)
After all, it is possible that the author is beta testing 454, just like the broad is. Usually, you get equipment very cheap, and in turn you put their name in your methods section. As far as I know, this sort of thing happens all the time, and it's good for both parties.
Was mike on the staff when they were doing the KTMA shows? You know, the ones that landed them the gig on comedy central? The ones that, without which, there would not have been any show for mike to be involved in?
Not that I don't like mike, mind you. They are two just different flavors of humor. Both are funny in different ways.
If programmers could write code ten times faster, that executes a tenth as quickly, that would actually be a beneficial trade-off for many (most?) organisations
Sounds perfect for prototyping. I'd love to have some code I could bang out to prototype a system, because my users often like to say "We'll know it's what we want when we see it." - and while you can say it's best to push the users harder, you'd be suprised at how well quick mockups work instead of taking huge amounts of time to spec on paper. Eventually, the mockups become to spec (or the use cases), then we can code. Doing that 10x faster would be fantastic.
Heck, at that point, users could play with the 'slow' version and fix workflow issues, etc, until the 'fast' version that supports the wider audience comes out.
We have a ton of anecdotes in which a mother knows when a child is in danger. However, we have zero anecdotes in which a father knows. This follows; the child spends 9 months in physical proximity to the mother, exchanging fluids; it's likely that entanglement is happening during that fluid exchange.
Not only that, but you have your mother's mitochondria. That's an organelle responsible for energy production that is not carried by sperm, and has it's own set of DNA.
I think Lucas misspelled it and called it 'midichlorians' or some pap like that...
In java (and hey, could be similar with RoR, etc), you write your web tier stuff to a middleware layer. There, we use Hibernate and HQL. That sort of crap 'just wouldn't parse' in HQL or criterion queries, as it would not be a valid parameter for "?".
Heck, even writing SQL, why aren't you writing parameterized prepared statements? Doesn't that also make you immune to this sort of crap?
"Select * from customers where Code = ?" setParam(1, myCraptasticUntextedVariable);// do this in your language of choice.
Or, am I missing something neat here? I like learning even more than being right...:)
Slashdot Mirror
I've had this discussion with my girlfriend many times before: when we want to stream HD content across the wireless network reliably from one room to another to support our video on demand system, we'll have to upgrade our wireless router.
Oh, you mean people have a need for bandwidth on the LAN, even though they're only connected to a 10Mb internet connection?
I'm in a different field, but...
I've had reviewers give me guff while reviewing population genetics papers because the reviewer didn't understand the basics of statistics. That's a field I shouldn't have to explain. If you can't grok a Chi-square test, then you should get out of the field, or at least not review papers.
You have to write for your audience, and assume some level of knowledge. You also often have to deal with word limits, so you can't write as much as you'd want to.
I have a friend who worked in the hedge fund game for a number of years. He's a brilliant mathematician, and worked on the models they used to inform their trading. The group he worked with was quite successful, and make a heck of a lot of money.
One of his most interesting comments: "The model can inform your decisions, but you have to know when to NOT trust the model." Another of his comments on a completely different talk: "Mathematical models are never perfect, but they can be useful."
The trading system can be modeled, but you can never capture all the true complexity of the real world. If you leave the model to do it's thing, if I know how it's going to act, I can game the system. If the world changes in a way that the model builders did not predict, then the system will also act inappropriately.
I can't imagine ever getting rid of all the traders out there, though I imagine expert systems will become more 'expert' as time goes on.
Where as on my version of firefox, it works just fine.
But then, I'm using windows so I should be the one with prob^H^H^H^H
Well, isn't that unusual?
Would the site having one single point of view make it more interesting to you? Or is this satire?
When you say there's nothing to parody by bad acting, you're referring to the movie Scream (1-4 is it now?), I assume. Or any other play on B movies (or you could parody bad writing with a novel like "Snow crash", which was just a parody of all the crappy cyperpunk before it.) My point is, "bad" gives plenty to parody. Generally with "much worse", but not with the seriousness.
(as an aside, I think people who create a bad but don't take themselves seriously are hard to parody, while it's people who think they're king of all they survey and turn out crap are easy targets...)
What Colbert does, is take on all of the bullshit attitude of some of the people you mention (especially O'Reilly), and say things that point out how rediculous the attitude is. Some examples from wikipedia:
The Colbert Report directly parodies The O'Reilly Factor with a commentary segment called "The Wørd", similar to O'Reilly's "Talking Points Memo". Like "the Memo", "The Wørd" features Colbert asserting a political point of view with a text screen graphic next to him. However, while O'Reilly's text serves to emphasize his points, Colbert's generally serve to provide an ironic counterpoint to his position. Also, Colbert's "No Fact Zone" is clearly inspired by O'Reilly's "No Spin Zone".
-------
Or, take for example the word "truthiness":
-------
In "The WØRD" on the first episode of the Report, Colbert featured the term "truthiness", which he defined as "The quality by which one purports to know something emotionally or instinctively, without regard to evidence or intellectual examination"
-------
If you don't find the attitude and over-the-top distortion amusing, then perhaps you just don't like the satire. I understand many people don't get satire (there's an interesting scientific study about satire and who does and doesn't get it that was published a few years ago), or perhaps it's just not your style. In either event, that's acceptable, and the above comments are not a flame of your preferences.
"So it comes to this: you cheer on Colbert because he "skewers" people with whom you disagree. "
You should watch O'Reilly, so you can see that Colbert is a paradody of the right wing "I'll shout louder than you" argumentation style. Heck, listen to talk radio for a little while, and you'll get the idea.
It's a parody of a style of 'conversation', where there's no dialog taking place, just spewing of the party line, louder than the other person. Besides, if you don't like what the other person is saying, you just shut their mike off.
Strangely (?) I read that you can fix this coil whine with a little superglue. Someone claimed that you could immobilize the coils so they didn't move, and it would not effect their operation.
Unfortunately, I don't rememeber if I read that here, or on the unofficial thinkpad forum, or somewhere else, so I don't have a reference. Perhaps someone with better hardware knowledge than I could confirm or deny this rumor.
That's an interesting set of comments.
My experience, while limited, is that if you pick the right company, then they will honor their warantee and fix the computer while it's in warantee. The question is: how long do you want to have a warantee for?
Dell fixed our laptops for 3 years. The laptops had many more problems than I would have liked (probably 12-15 repairs on 5 laptops over 3 years.) Now that the warantee on them is over, we'll replace them as they die with something that is in warantee.
The new thinkpads we're buying all have 3 year complete care warantees. I dropped a glass of water in one, and it's being fixed for free. I expect we'll get 3 years out of those laptops, and any extra time is a bonus.
I think the question of warantees that is important is: How reasonable are the service people, and how much work do you have to do to get your computer serviced? Dell makes you jump through hoops, and they seem to screw up our server more than they fix it. Lenovo seems to take you at your word, and fix things ASAP. I must admit that I also like the fact that lenovo's phone support has enough people staffing it that you get through quickly, and the call centers are in the country I'm in (the US), so I have no communications gap. Dell seems to do most of their support over IRC-style chat, or email, and the gaps in communication are much wider.
Holy Crap. I lived in Cary, NC for a year. I couldn't get out quick enough. Sure, good salary, and good housing. No nightlife, no good food, you had to drive 10-20 miles to get anywhere. I'm originally from NY, and that's not my idea of a 'city'. Boston suits me much better, and while 'biotech' is big in RTP, the largest research centers are in Boston/San Fan. RTP seems to have as much manufacturing as research - which makes it sound like a 'center', but there are not nearly as many PhD's in the area as you'd think.
I'd have to wonder if any of the other areas are any better.
Plus, I'd hate to have to mail order all my food. I love to eat, and large cities tend to have resturaunts and grocery stores for immigrant populations. I love to go to the local brazillian, or japapese, or korean, or pakistani grocery store to buy all those ingredients you can't get at a chain store.
I think a lot of smaller cities offer good ratio of salary to housing, but what if you need more?
Actually, in many cases, the 'pure luck' option now is actually 'test all possible causes of type '. For example, if I'm going to study Alhemiers with the intent of finding the genetic variation responsible for the disease, I don't want to read literature papers, pick genes *I* think are involved, and just look at those areas. The emerging trend now is to test 95%+ of all common variation in the human genetic sequence, and find the variation that tracks with the disease. You then work *back* to functional information about how those variations affect the biology.
I'd say "understanding how the brain works" is, for some kinds of science, the old school method. Why? Because our understanding is incomplete, our tests using best guesses haven't yielded a lot of luck, so why not explore the whole problem space? Actually, as one of my well known colleges says "If there's one thing we've found out over the last 10 years, it's that we're exceptionally bad at picking regions to study."
You need to re-read this bit: Maybe you should have read TFA more carefully, as it in fact gives two specific cases of utility.
They *tell* you it's useful, but they don't supply the information to demonstrate that it's useful.
I can tell you about this great bridge I have to sell you. Can you tell me why you'd want to listen to me and give me money?
I think putting up graphs without even the kindness of a legend, labels, etc is not terribly useful. And while I'm not in the protein folding field, I'm close by in genetics, and we look at protein folding and how mutations affect active sites.
You need to re-read this bit:
but if they do not convey information (and not lose a large amount of relevant information) then they are just a nice way to generate patterns for some nerd's tie.
Like I said in another post, these guys ought to read a bit of Tufte, so the graphs are usable. Unless of course this article blatently cuts and pastes only select parts of the graph that look "pretty". In which case, the authors of the article are the people who have rendered them useless.
Thanks for letting me know what I don't do for a living, though, as I'd nearly forgotten and thought I was a professional chef.
When I read articles about biology, especially molecular biology/genetics, I see lots of interesting "facts" about the field given by various members of the slashdot crowd. I'm not a leader in the field, but I feel knowledgeable enough working in the field to know just how wrong these "facts" are, yet get modded insightful.
What scares me are all the articles about topics that I'm not an expert in, where I can't judge the veracity of comments. I've realized that if you guys are so terribly wrong here, that you're probably not believeable anywhere else, either.
Not that this news to anyone. It just depresses me everytime I see this type of story come up.
*sigh*
I don't buy junk DNA. The reason? There's lots of regions that are not in coding regions, but are conserved (read: the same) with other organisms much higher than chance would predict. These regions have been tested, and found that they are being kept the same under via selection.
Want to know more? Read Nat Genet. 2006 Feb;38(2):223-7.
In the interest of full disclosure: I'm one of the authors, and I'm sick of reading about 'junk dna'.
The thing is, if it's very hard to communicate the information without training to interpret it, then you'd probably want to go back and refactor the design to increause the usability. Does no one read Tufte anymore?
I haven't had to cast an object since I started using java 1.5 18 months ago. Generics, look into them. Also, you can put primitives into collections (via autoboxing).
Java 1.5 made the language a LOT nicer and more expressive.
This is exactly right. This is why there's a methods/materials section in papers, so that we have the reagents and the process neccesary to reproduce the results you put in the discussion section.
The point of visualizing data is to learn something that you could not do with the raw data. In all of the cases shown in the article (yes, I acually read TFA), I didn't spot an example where it actually showed anything useful.
The first example with proteins: how similar are two proteins? If two shapes are similar (and please, how many proteins where being graphed there? One, two, five?), then you might be able to recognize it. If they are similar shapes, are they always presented in the same orientation in space? Does color have any meaning? Does this graph have any legend? If I gave someone who understood the graphs two proteins, what could he say besides "these are related" and "these are not related"? We already have wonderful programs to compare two proteins and say how similar they are two each other, along with being able to the estimate significance of the measurement.
I'm not sure that the other graphics look more informative. They are all pretty, but if they do not convey information (and not lose a large amount of relevant information), then they are just a nice way to generate patterns for some nerd's tie.
I put a lot of stock in the fact that some politicians can talk, and express their ideas, instead of just releasing press kits.
Listen to bush some time, and then read his press kits. While he's trying desperately to repeat what's in the press kits (until he goes 'off script' and makes a fool of himself), other officials have a lot more elegance and charisma. Listen to Barack OBama speak - that man has passion, intensity, and intelligence. That's the sort of guy I want out in the world, trying to improve our foreign policy.
The Broad institute has 454 as one of their sequencing platforms in house. Of course, george church is working on an 'open source' sequencing platform that will be obscenely cheap by comparison. I think you need a microscope, some (comparably) reagents, and some open source software to do sequencing.
You might remember the Broad institute when they were the whitehead center for genome research - the people who sequenced a large (the largest?) part of the human genome. A number of people there seem to think that the 454 technology is at least 'pretty interesting'.
I'm interesting in anyone who can do the $1000/person sequencing. We're using Affy 500K chips now to do whole genome scans, but why not get the full sequence (and find those out of LD, really rare SNPs as well as indels...)
After all, it is possible that the author is beta testing 454, just like the broad is. Usually, you get equipment very cheap, and in turn you put their name in your methods section. As far as I know, this sort of thing happens all the time, and it's good for both parties.
Was mike on the staff when they were doing the KTMA shows? You know, the ones that landed them the gig on comedy central? The ones that, without which, there would not have been any show for mike to be involved in?
Not that I don't like mike, mind you. They are two just different flavors of humor. Both are funny in different ways.
Sounds perfect for prototyping. I'd love to have some code I could bang out to prototype a system, because my users often like to say "We'll know it's what we want when we see it." - and while you can say it's best to push the users harder, you'd be suprised at how well quick mockups work instead of taking huge amounts of time to spec on paper. Eventually, the mockups become to spec (or the use cases), then we can code. Doing that 10x faster would be fantastic.
Heck, at that point, users could play with the 'slow' version and fix workflow issues, etc, until the 'fast' version that supports the wider audience comes out.
Not only that, but you have your mother's mitochondria. That's an organelle responsible for energy production that is not carried by sperm, and has it's own set of DNA.
I think Lucas misspelled it and called it 'midichlorians' or some pap like that...
Who even writes SQL these days directly?
// do this in your language of choice.
In java (and hey, could be similar with RoR, etc), you write your web tier stuff to a middleware layer. There, we use Hibernate and HQL. That sort of crap 'just wouldn't parse' in HQL or criterion queries, as it would not be a valid parameter for "?".
Heck, even writing SQL, why aren't you writing parameterized prepared statements? Doesn't that also make you immune to this sort of crap?
"Select * from customers where Code = ?"
setParam(1, myCraptasticUntextedVariable);
Or, am I missing something neat here? I like learning even more than being right...:)