Reading up and knowing what you're getting into beforehand is necessary
why ? If "they " wanted to , they could make easy to use DBs, or at least easy to use lite DBs.
This sort of geek arrogant attutide, that the user must conform to a procrustean bed of software, is what holds back linux and so much other software from wide acceptance.
One of the things that makes MS dominant is that they assume the user is stupid, and make it easy
you are ignorant and bigoted, and not only that, you believe in this ancient fairy tale as the word of god for two long liberals have put up with you racist bigoted creeps who use the bible as an excuse for hatred
thinks "premium" popcorn is morally revolting ? It is nothing but marketing garbage. You can buy a huge jar of plain popcorn for a buck, or be brainwashed and buy fancy stuff for much more.
There is something sick and evil about a country which thinks spending energy on marketing popcorn is a reasonable way to make a living.
to reply to anything having to do with the original article, whichis incohernet and poorly written. Tom Yager couldnt explain how to pound sand down a rathole without pictures, much less can he say anything usefull about a complex subject like market share for competing computer systems.
Computers and cars R very similar, in that there is a limited number of suppliers, and the cost of the item is , to a very large extent, driven by volume and automated manufacturing.
If any of the posters in this thread had the slightest idea of how things are actually made, they would know that a non standard item is always more expsenseive.
There are always a small minority bitching that they cant get their pet item. I don;t know what the analogy in car land is, but I am sure there are a lot of car buffs pissed off because what they want is not sold by std vendors, and is an expspensive aftermarket addon.
I don't understand your logic as to why software makes a product different. You still buy it at the store, you still look for a cheap item, etc. What does the general purpose thing have to do with how I go to compusa or whatever and look at the 599, the 699, the 799, etc - this is exactly like going to toyota, and looking at the camry ce, the le, the xle, leather.... I guess the exact analogy would be ethanol fueled cars, or something like that.
I note that the process involves removing blood from the body, running the process on cells purified from teh blood, and then reinjecting the cells back intothe patient.
I believe, but am not an expert in this field, that the simple process of removing cells from the body is in and of itself, highly mutagenic.
There is *no way* the cells could be checked before reinjection.
In any event, it is an interesting piece of science, but a LONG way from clinical practice - stay tuned for the update in 2020.
I have not read the article, but repair processes can be "error prone". That is, the mechanisms cells use to repair DNA often involve high error rates.
The human genome is 3e9 BP long (roughly..not counting indels, the unsequenced centromeres, etc etc)
So the chemical process of identifying the one single mutated basepair has to have a chemical specificity of >>1e9, because there are >>1e6 cells that are exsposed. That is, lets say you feed the reagent to a person. Millions of cells, each with 1e9 bp, are expsosed. Say the process has an error rate of 1e10 - many, many cells will have incorrect repairs done This is just like error rates in, say, reading data from a harddrive: the larger the file, the lower the error rte has to be
What/.ers may not appreciate is that typically, it is VERY, repeat VERY hard to get chemcial reaction specificity of anywhere close to 1e9 for reactions invovling DNA.
at least on windows2000, there are many files vlc shd play but does not, not to mention the hangs and freezes adn need for reinstalls (actually, sadly, like most programs, vlc has gotten worse with time)
There is a reason most of bioniformatics is simply DB and looking at the data, and that is because that is what is reasonable.
First, there has been a huge, huge explosion of data, and the bio community was really not prepared, and so simply getting an understanding of what a real DB is, and how to set them up and so forth, took a while. Let me tell you a true story: This guy tells me, I used to work on parvovirus ( its not important what parvoviruses do)and I am looking in gen bank, and this PV seq is in the database backwards - its obvious. So I call Genbank up, and they say, the rules are, only the original submitter can make a change. So here we have THE major public database of DNA sequence, and there is no way to flag obvious errors ( i mean raelly ovbious errors like tranposing letters) So, there has been a painful period where the rather small and insular bio community learned about data storage. The other, more significant problem, is that we really dont know what to do with all the data - that is, the bottleneck is in finding scientists who can say, it would be interesting to compare this to this, or do this sort of analysis, and who can then write the code. Because a lot of the data we have has no biological function - it is just DNA sequence, which in and of itself is as informative as a page of code without any knowledge of the hardware or OS that run the code.
Another problem is that a lot of the intersting problems are known to require ~ pflop computational power, that is, we know there is a class of very important problems, and (from what i gather) all the experts agree we need ~ 1,000 x more flops (ibms blue gene is supposed to be the first machine to break this barrier.)
If I had to advise you, I would say in the next 5 years, far and away the most interesting place in bioinformatics is in protein folding and drug docking. Proteins are linear strings of 20 chemical building blocks, eg ala gly ala is a protein with three blocks, alanine, glycine, alanine. most real proteins have >200 strings, and the initial linear molecule folds up into a 3D structure. Currently, the cost of obtaining a 3D structure (x ray crystallography structure, or 2D NMR) typically exceeds the cost of obtaining the linear (sequence) info by two to orders of magnitude. since it is the 3D structure that is biologically relevant, there is tremendous interest in programs that can predict 3D structure from 1D sequence.
The docking problem is, once you have a 3D structure for your protein, if you are given the 3D structure of 10,000 possible drugs, can you get the computer to tell you which of those drugs binds (key/lock fit) to the protein...whoever can do this will truly revolutionize (I mean really) the drug industry...
another formidable problem is, once you know all the proteins in the human body, can you predict what happens when you raise the concentration in blood of one protein ? And how do genetic variations affeect this process.
the markets are very small, and there are a lot of companies in each market, so they cant actually afford any real programmers to write real algorythms; instead, they take some piece of code written by a grad student , add a fancy but not very usable gui, and sell the result..
We currently double glove, with nitrile over latex, but I hear that gloves made from diamond coated (CVD, chemical vapor deposistion) BeO (berylium oxide) reinforced PEEK (polyetheretherketone) work really well.
in theory, the qdots ar more stable (less photobleaching) a recognized problem with std labels, and they have narrower emission spectra, so multiplexing is easier (eg std labels like fluorescein and rhodamine have wide emission spectra that overlap)(altho the lanthanide chelates have 10 nm fwhm) potentially, you can tune the excitation and emission spectra to match your laser lines, so if someone develops a real cheap stable diode laser, you can tune the dot to that line on the other hand, the qdots are big enough that they might not get to all the relevant sites.
neither the parent nor the first reponse are correct if you look at single cell organisms, which "normally" live in, say pond water, you can examine them in pretty close to thier normal env. Or say, a sperm cell - that exists outside the body. And every scientist is painfully aware that many cells are not normal outside the body; there are whole books on this Also, it is well established that you need a tag to look at, say cell surface proteins; this is done everyday. Not sure how nanoprobes help map a proteins path thru a membrane - sounds like gibberish. if you mean can it help map the route a protein takes thru pores, such as when a protein transits from cytoplasmic ribosomes to the nucleus, GFP fusions already do this very well
See, for instance, the quantum dot company (www.qdot.com). What is new is using a bio tag to direct the dot into the nucleus; such tags ("nuclear localiztion signals") are well known in theliteratrue for proteins, so what is new is that they took qdots and coated them with one of these signals. So, this is an addittion to the large catalog of optical probes that biologiest have.
like everyone else i use google a lot, and compared to what i think it shd do, it sucks.
of course, that is an obkective std, but I wd say people are doing an MS here- they accept the horrible flaws in google as normal, the way non/. people accept the horrible flaws of MS as normal.
there are so many wyays google cd be so much better, it is scary.
no reason at all apologize dumbing down - in fact, the first post on thematter that made sense, and the first poster with enough intelligence to explain it properly.
simple is NOT stupid - simple is smart, or as Einstein said, any scientist who could not explain himself to a milkmaid did not understand what he was working on OR
This is beside the point - niether bankrate.com (which does not really present cd rates in a usable form), or the much better money-rates.com or ibankdesign.com is the equivalent of open source, and it shows
I disagree with your financial analysis also, but that is another discussion.
This seems like an un explored area - surely many of hte people who use the wiki would like a table of the best cd rates , updated, say daily.
The problem of how to compile a list of CD rates is formidable, as there are hundreds, if not thousands of banks (most of which almost alwyas have low rates) not all of which have online rates, and there are "specials"
As exactly the sort of user u describe, all I can say is that for some reason I have gotten the impression that *BSD is out of date, not GUI based (kiss of death) and has no hardware support or user base.
I'm not saying this is true or false: this is just the impression i have picked up from sites like/.
Also, when I go down to the store, there is a large section on Linux OS, and (I think) nothing on *BSD; I would suspect that a lot of cherry MS to *nix people like myself are willing to pay for a commercial box the first time
Also, I would imagine alot of us older users remeber *nix boxes from the prime/apollo/data general days, and boy, did they suck (I know, it is completely unfair to compair any electronic device from 15 years ago to today, but who said life was fair??)
Slashdot Mirror
I thought inflation preceeded all star formation, so how can the source be a pre inflationary epoch galaxy ?
Reading up and knowing what you're getting into beforehand is necessary
why ?
If "they " wanted to , they could make easy to use DBs, or at least easy to use lite DBs.
This sort of geek arrogant attutide, that the user must conform to a procrustean bed of software, is what holds back linux and so much other software from wide acceptance.
One of the things that makes MS dominant is that they assume the user is stupid, and make it easy
maybe if the DBs were a little easier to use.... always easy to blame the user, isnt it.
I once tried to download Mysql, just for laughs. Boy, those people could use a trainign lesson in how to present software to newbys....
you are ignorant and bigoted, and not only that, you believe in this ancient fairy tale as the word of god for two long liberals have put up with you racist bigoted creeps who use the bible as an excuse for hatred
thinks "premium" popcorn is morally revolting ? It is nothing but marketing garbage. You can buy a huge jar of plain popcorn for a buck, or be brainwashed and buy fancy stuff for much more. There is something sick and evil about a country which thinks spending energy on marketing popcorn is a reasonable way to make a living.
Wow - u qualify as a slashdot uber expert
to reply to anything having to do with the original article, whichis incohernet and poorly written.
Tom Yager couldnt explain how to pound sand down a rathole without pictures, much less can he say anything usefull about a complex subject like market share for competing computer systems.
Computers and cars R very similar, in that there is a limited number of suppliers, and the cost of the item is , to a very large extent, driven by volume and automated manufacturing.
If any of the posters in this thread had the slightest idea of how things are actually made, they would know that a non standard item is always more expsenseive.
There are always a small minority bitching that they cant get their pet item. I don;t know what the analogy in car land is, but I am sure there are a lot of car buffs pissed off because what they want is not sold by std vendors, and is an expspensive aftermarket addon.
I don't understand your logic as to why software makes a product different. You still buy it at the store, you still look for a cheap item, etc. What does the general purpose thing have to do with how I go to compusa or whatever and look at the 599, the 699, the 799, etc - this is exactly like going to toyota, and looking at the camry ce, the le, the xle, leather....
I guess the exact analogy would be ethanol fueled cars, or something like that.
I note that the process involves removing blood from the body, running the process on cells purified from teh blood, and then reinjecting the cells back intothe patient.
I believe, but am not an expert in this field, that the simple process of removing cells from the body is in and of itself, highly mutagenic.
There is *no way* the cells could be checked before reinjection.
In any event, it is an interesting piece of science, but a LONG way from clinical practice - stay tuned for the update in 2020.
I have not read the article, but repair processes can be "error prone". That is, the mechanisms cells use to repair DNA often involve high error rates.
/.ers may not appreciate is that typically, it is VERY, repeat VERY hard to get chemcial reaction specificity of anywhere close to 1e9 for reactions invovling DNA.
The human genome is 3e9 BP long (roughly..not counting indels, the unsequenced centromeres, etc etc)
So the chemical process of identifying the one single mutated basepair has to have a chemical specificity of >>1e9, because there are >>1e6 cells that are exsposed. That is, lets say you feed the reagent to a person. Millions of cells, each with 1e9 bp, are expsosed. Say the process has an error rate of 1e10 - many, many cells will have incorrect repairs done
This is just like error rates in, say, reading data from a harddrive: the larger the file, the lower the error rte has to be
What
I will rtfa,
at least on windows2000, there are many files vlc shd play but does not, not to mention the hangs and freezes adn need for reinstalls (actually, sadly, like most programs, vlc has gotten worse with time)
There is a reason most of bioniformatics is simply DB and looking at the data, and that is because that is what is reasonable.
First, there has been a huge, huge explosion of data, and the bio community was really not prepared, and so simply getting an understanding of what a real DB is, and how to set them up and so forth, took a while.
Let me tell you a true story: This guy tells me, I used to work on parvovirus ( its not important what parvoviruses do)and I am looking in gen bank, and this PV seq is in the database backwards - its obvious. So I call Genbank up, and they say, the rules are, only the original submitter can make a change.
So here we have THE major public database of DNA sequence, and there is no way to flag obvious errors ( i mean raelly ovbious errors like tranposing letters)
So, there has been a painful period where the rather small and insular bio community learned about data storage.
The other, more significant problem, is that we really dont know what to do with all the data - that is, the bottleneck is in finding scientists who can say, it would be interesting to compare this to this, or do this sort of analysis, and who can then write the code. Because a lot of the data we have has no biological function - it is just DNA sequence, which in and of itself is as informative as a page of code without any knowledge of the hardware or OS that run the code.
Another problem is that a lot of the intersting problems are known to require ~ pflop computational power, that is, we know there is a class of very important problems, and (from what i gather) all the experts agree we need ~ 1,000 x more flops (ibms blue gene is supposed to be the first machine to break this barrier.)
If I had to advise you, I would say in the next 5 years, far and away the most interesting place in bioinformatics is in protein folding and drug docking. Proteins are linear strings of 20 chemical building blocks, eg ala gly ala is a protein with three blocks, alanine, glycine, alanine.
most real proteins have >200 strings, and the initial linear molecule folds up into a 3D structure. Currently, the cost of obtaining a 3D structure (x ray crystallography structure, or 2D NMR) typically exceeds the cost of obtaining the linear (sequence) info by two to orders of magnitude.
since it is the 3D structure that is biologically relevant, there is tremendous interest in programs that can predict 3D structure from 1D sequence.
The docking problem is, once you have a 3D structure for your protein, if you are given the 3D structure of 10,000 possible drugs, can you get the computer to tell you which of those drugs binds (key/lock fit) to the protein...whoever can do this will truly revolutionize (I mean really) the drug industry...
another formidable problem is, once you know all the proteins in the human body, can you predict what happens when you raise the concentration in blood of one protein ?
And how do genetic variations affeect this process.
the markets are very small, and there are a lot of companies in each market, so they cant actually afford any real programmers to write real algorythms; instead, they take some piece of code written by a grad student , add a fancy but not very usable gui, and sell the result..
We currently double glove, with nitrile over latex, but I hear that gloves made from diamond coated (CVD, chemical vapor deposistion) BeO (berylium oxide) reinforced PEEK (polyetheretherketone) work really well.
in theory, the qdots ar more stable (less photobleaching) a recognized problem with std labels, and they have narrower emission spectra, so multiplexing is easier (eg std labels like fluorescein and rhodamine have wide emission spectra that overlap)(altho the lanthanide chelates have 10 nm fwhm)
potentially, you can tune the excitation and emission spectra to match your laser lines, so if someone develops a real cheap stable diode laser, you can tune the dot to that line
on the other hand, the qdots are big enough that they might not get to all the relevant sites.
neither the parent nor the first reponse are correct
if you look at single cell organisms, which "normally" live in, say pond water, you can examine them in pretty close to thier normal env. Or say, a sperm cell - that exists outside the body. And every scientist is painfully aware that many cells are not normal outside the body; there are whole books on this
Also, it is well established that you need a tag to look at, say cell surface proteins; this is done everyday. Not sure how nanoprobes help map a proteins path thru a membrane - sounds like gibberish. if you mean can it help map the route a protein takes thru pores, such as when a protein transits from cytoplasmic ribosomes to the nucleus, GFP fusions already do this very well
See, for instance, the quantum dot company (www.qdot.com). What is new is using a bio tag to direct the dot into the nucleus; such tags ("nuclear localiztion signals") are well known in theliteratrue for proteins, so what is new is that they took qdots and coated them with one of these signals. So, this is an addittion to the large catalog of optical probes that biologiest have.
like everyone else i use google a lot, and compared to what i think it shd do, it sucks.
/. people accept the horrible flaws of MS as normal.
of course, that is an obkective std, but I wd say people are doing an MS here- they accept the horrible flaws in google as normal, the way non
there are so many wyays google cd be so much better, it is scary.
I think it is called marketing and its goal is to create buzz...and /. bought it, hoook, line and sinker
All those horrid things those nasty morse code people do ...
no reason at all apologize dumbing down - in fact, the first post on thematter that made sense, and the first poster with enough intelligence to explain it properly.
simple is NOT stupid - simple is smart, or as Einstein said, any scientist who could not explain himself to a milkmaid did not understand what he was working on
OR
don't look entirely convincing, espcially in the second case
s c2 005-09/ssc2005-09a.shtml
http://www.spitzer.caltech.edu/Media/releases/s
did anyone see how they discount things like solar flares
This is beside the point - niether bankrate.com (which does not really present cd rates in a usable form), or the much better money-rates.com or ibankdesign.com is the equivalent of open source, and it shows
I disagree with your financial analysis also, but that is another discussion.
This seems like an un explored area - surely many of hte people who use the wiki would like a table of the best cd rates , updated, say daily. The problem of how to compile a list of CD rates is formidable, as there are hundreds, if not thousands of banks (most of which almost alwyas have low rates) not all of which have online rates, and there are "specials"
..impossible to explain ...*BSDs
/.
As exactly the sort of user u describe, all I can say is that for some reason I have gotten the impression that *BSD is out of date, not GUI based (kiss of death) and has no hardware support or user base.
I'm not saying this is true or false: this is just the impression i have picked up from sites like
Also, when I go down to the store, there is a large section on Linux OS, and (I think) nothing on *BSD; I would suspect that a lot of cherry MS to *nix people like myself are willing to pay for a commercial box the first time
Also, I would imagine alot of us older users remeber *nix boxes from the prime/apollo/data general days, and boy, did they suck (I know, it is completely unfair to compair any electronic device from 15 years ago to today, but who said life was fair??)