1605: "DNA"

This forum is for the individual discussion thread that goes with each new comic.

Moderators: Moderators General, Prelates, Magistrates

lorb
Posts: 405
Joined: Wed Nov 10, 2010 10:34 am UTC
Location: Austria

1605: "DNA"

Postby lorb » Wed Nov 18, 2015 2:20 pm UTC

Image

Title Text: "Researchers just found the gene responsible for mistakenly thinking we've found the gene for specific things. It's the region between the start and the end of every chromosome, plus a few segments in our mitochondria."

Actually the source of google.com is not so bad once properly formatted. Put in a lot of newlines and spaces and replace one-character variable names with something nice and it starts to make sense.
Please be gracious in judging my english. (I am not a native speaker/writer.)
http://decodedarfur.org/

commodorejohn
Posts: 1176
Joined: Thu Dec 10, 2009 6:21 pm UTC
Location: Placerville, CA
Contact:

Re: 1605: "DNA"

Postby commodorejohn » Wed Nov 18, 2015 2:29 pm UTC

I know so many people like this...
"'Legacy code' often differs from its suggested alternative by actually working and scaling."
- Bjarne Stroustrup
www.commodorejohn.com - in case you were wondering, which you probably weren't.

eidako
Posts: 126
Joined: Wed Apr 06, 2011 10:24 am UTC

Re: 1605: "DNA"

Postby eidako » Wed Nov 18, 2015 2:38 pm UTC

Well, at least they still have cuttlefish.

User avatar
Whizbang
The Best Reporter
Posts: 2238
Joined: Fri Apr 06, 2012 7:50 pm UTC
Location: New Hampshire, USA

Re: 1605: "DNA"

Postby Whizbang » Wed Nov 18, 2015 2:45 pm UTC

Maybe we can invent a genetic algorithm to read it for us...

User avatar
The Moomin
Posts: 358
Joined: Wed Oct 13, 2010 6:59 am UTC
Location: Yorkshire

Re: 1605: "DNA"

Postby The Moomin » Wed Nov 18, 2015 3:31 pm UTC

Isn't everything about human DNA contained in the appendix?

We can extrapolate from this to other animals.

Then I don't know what.
I'm alive because the cats are alive.
The cats are alive because I'm alive.
Specious.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: 1605: "DNA"

Postby Izawwlgood » Wed Nov 18, 2015 3:38 pm UTC

commodorejohn wrote:I know so many people like this...
SO MANY PEOPLE
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

jgh
Posts: 127
Joined: Thu Feb 03, 2011 1:04 pm UTC

Re: 1605: "DNA"

Postby jgh » Wed Nov 18, 2015 3:50 pm UTC

lorb wrote:Actually the source of google.com is not so bad once properly formatted. Put in a lot of newlines and spaces and replace one-character variable names with something nice and it starts to make sense.

Blooody 'ell! Why do *we* need that? It should be done at the Google end, all we need is title, img src=pretty picture, form=text box, button=submit. The Google search box on my home page is 367 bytes including pretty formatting.

User avatar
Envelope Generator
Posts: 582
Joined: Sat Mar 03, 2012 8:07 am UTC
Location: pareidolia

Re: 1605: "DNA"

Postby Envelope Generator » Wed Nov 18, 2015 3:59 pm UTC

The Moomin wrote:Isn't everything about human DNA contained in the appendix?

We can extrapolate from this to other animals.

Then I don't know what.


Why is the important stuff always relegated to an appendix.
I'm going to step off the LEM now... here we are, Pismo Beach and all the clams we can eat

eSOANEM wrote:If Fonzie's on the order of 100 zeptokelvin, I think he has bigger problems than difracting through doors.

ayryq
Posts: 12
Joined: Tue Feb 12, 2013 12:48 pm UTC

Re: 1605: "DNA"

Postby ayryq » Wed Nov 18, 2015 4:07 pm UTC

Well, I tried to post the current google source code in a reply, but I couldn't manage it. Too many characters, plus it gets flagged as spam.

User avatar
HES
Posts: 4888
Joined: Fri May 10, 2013 7:13 pm UTC
Location: England

Re: 1605: "DNA"

Postby HES » Wed Nov 18, 2015 4:13 pm UTC

At 213 thousand characters, it's about four times over the post limit. I'm sure people are capable of the three or so clicks required to see it for themselves.
He/Him/His Image

User avatar
Whizbang
The Best Reporter
Posts: 2238
Joined: Fri Apr 06, 2012 7:50 pm UTC
Location: New Hampshire, USA

Re: 1605: "DNA"

Postby Whizbang » Wed Nov 18, 2015 4:14 pm UTC

ayryq wrote:Well, I tried to post the current google source code in a reply, but I couldn't manage it. Too many characters, plus it gets flagged as spam.


I am glad you shared this with the group. Plenty of food for thought and discussion.

So, question: Optimization by humans appears to be quicker than optimization that occurs naturally (artificial selection vs natural selection), at least for specific vectors and desired outcomes. Would there come a point where genetic algorithms result in a more complex system than a DNA based lifeform? It has been stated by people smarter than me that the only way to simulate how a particular DNA strand would develop into a living creature would be in real-time in a real environment (ie the only way to see how a person would grow is to grow a person). Will this always be true?

ayryq
Posts: 12
Joined: Tue Feb 12, 2013 12:48 pm UTC

Re: 1605: "DNA"

Postby ayryq » Wed Nov 18, 2015 4:15 pm UTC

You're welcome. I failed lots of different ways too. Probably the reason I posted anyhow is genetic.

fatcatfan
Posts: 3
Joined: Wed Nov 18, 2015 4:12 pm UTC

Re: 1605: "DNA"

Postby fatcatfan » Wed Nov 18, 2015 4:17 pm UTC

lorb wrote:Put in a lot of newlines and spaces and replace one-character variable names with something nice and it starts to make sense.


Exactly, it's optimised for bandwidth. I'm confident there's an unoptimised version google uses for maintaining the code.

User avatar
Neil_Boekend
Posts: 3220
Joined: Fri Mar 01, 2013 6:35 am UTC
Location: Yes.

Re: 1605: "DNA"

Postby Neil_Boekend » Wed Nov 18, 2015 4:21 pm UTC

Whizbang wrote:
ayryq wrote:Well, I tried to post the current google source code in a reply, but I couldn't manage it. Too many characters, plus it gets flagged as spam.


I am glad you shared this with the group. Plenty of food for thought and discussion.

So, question: Optimization by humans appears to be quicker than optimization that occurs naturally (artificial selection vs natural selection), at least for specific vectors and desired outcomes. Would there come a point where genetic algorithms result in a more complex system than a DNA based lifeform? It has been stated by people smarter than me that the only way to simulate how a particular DNA strand would develop into a living creature would be in real-time in a real environment (ie the only way to see how a person would grow is to grow a person). Will this always be true?

If you naively assume that the computational power available for a simulation keeps increasing as it has done for decades then the answer is probably no. We can simulate molecular reactions, just not enough to simulate a whole human cell, let alone a complete human.
Mikeski wrote:A "What If" update is never late. Nor is it early. It is posted precisely when it should be.

patzer's signature wrote:
flicky1991 wrote:I'm being quoted too much!

he/him/his

ayryq
Posts: 12
Joined: Tue Feb 12, 2013 12:48 pm UTC

Re: 1605: "DNA"

Postby ayryq » Wed Nov 18, 2015 4:34 pm UTC


MacTown
Posts: 1
Joined: Wed Nov 18, 2015 4:26 pm UTC

Re: 1605: "DNA"

Postby MacTown » Wed Nov 18, 2015 4:39 pm UTC

If DNA is source code, life is written in Malbolge

commodorejohn
Posts: 1176
Joined: Thu Dec 10, 2009 6:21 pm UTC
Location: Placerville, CA
Contact:

Re: 1605: "DNA"

Postby commodorejohn » Wed Nov 18, 2015 4:58 pm UTC

Neil_Boekend wrote:If you naively assume that the computational power available for a simulation keeps increasing as it has done for decades then the answer is probably no.

Emphasis on "naive." Karl Schwarzchild might have a few things to say about the notion that we can just continue to pack arbitrary amounts of computer into increasingly smaller spaces forever.
"'Legacy code' often differs from its suggested alternative by actually working and scaling."
- Bjarne Stroustrup
www.commodorejohn.com - in case you were wondering, which you probably weren't.

User avatar
cellocgw
Posts: 2043
Joined: Sat Jun 21, 2008 7:40 pm UTC

Re: 1605: "DNA"

Postby cellocgw » Wed Nov 18, 2015 5:00 pm UTC

Yeah, but....

How much ya wanna bet nobody ever put comments into the DNA source code? :P

Plus we still haven't figured out what the comment/ignore_this_part character is in DNA. It seems amazingly capable of activating and deactivating all sorts of unexpected subroutines. First DNA makes proteins and other stuff come into being, then those molecules (some of them) come back and modify the source code (DNA), which then goes off and makes other things happen or not happen. Clearly there's no SVN or release control here!
https://app.box.com/witthoftresume
Former OTTer
Vote cellocgw for President 2020. #ScienceintheWhiteHouse http://cellocgw.wordpress.com
"The Planck length is 3.81779e-33 picas." -- keithl
" Earth weighs almost exactly π milliJupiters" -- what-if #146, note 7

User avatar
Neil_Boekend
Posts: 3220
Joined: Fri Mar 01, 2013 6:35 am UTC
Location: Yes.

Re: 1605: "DNA"

Postby Neil_Boekend » Wed Nov 18, 2015 5:01 pm UTC

commodorejohn wrote:
Neil_Boekend wrote:If you naively assume that the computational power available for a simulation keeps increasing as it has done for decades then the answer is probably no.

Emphasis on "naive." Karl Schwarzchild might have a few things to say about the notion that we can just continue to pack arbitrary amounts of computer into increasingly smaller spaces forever.

Only economists and idiots believe in unlimited expansion in a limited space.
Mikeski wrote:A "What If" update is never late. Nor is it early. It is posted precisely when it should be.

patzer's signature wrote:
flicky1991 wrote:I'm being quoted too much!

he/him/his

User avatar
Copper Bezel
Posts: 2426
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: 1605: "DNA"

Postby Copper Bezel » Wed Nov 18, 2015 5:08 pm UTC

ayryq wrote:How about a picture of the source code :)
https://www.dropbox.com/s/gtbzqei780hcyq0/googlesource.png?dl=0


Seeing that narrow light-gray column and wondering whether something had gone awry, and then realizing what I was looking at, was well worth the click.

Whizbang wrote:So, question: Optimization by humans appears to be quicker than optimization that occurs naturally (artificial selection vs natural selection), at least for specific vectors and desired outcomes. Would there come a point where genetic algorithms result in a more complex system than a DNA based lifeform? It has been stated by people smarter than me that the only way to simulate how a particular DNA strand would develop into a living creature would be in real-time in a real environment (ie the only way to see how a person would grow is to grow a person). Will this always be true?


I like the idea. I suppose that it wouldn't be as impenetrable to the researchers who created it as biology is to the people studying it now, but I'd tend to expect that by the time we had some system of nested genetic algorithms doing their thing quite that autonomously, we'd probably also separately have a stronger handle on biology, so.

cellocgw wrote:Yeah, but....

How much ya wanna bet nobody ever put comments into the DNA source code? :P


Why continue the source code metaphor? DNA is written compiled.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: 1605: "DNA"

Postby Izawwlgood » Wed Nov 18, 2015 5:16 pm UTC

cellocgw wrote:How much ya wanna bet nobody ever put comments into the DNA source code? :P

Plus we still haven't figured out what the comment/ignore_this_part character is in DNA. It seems amazingly capable of activating and deactivating all sorts of unexpected subroutines. First DNA makes proteins and other stuff come into being, then those molecules (some of them) come back and modify the source code (DNA), which then goes off and makes other things happen or not happen. Clearly there's no SVN or release control here!

Actually, 'commenting out' bits of code could reasonably be analogous to what transposons of various sorts do.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
orthogon
Posts: 3057
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1605: "DNA"

Postby orthogon » Wed Nov 18, 2015 5:41 pm UTC

Copper Bezel wrote:
Whizbang wrote:So, question: Optimization by humans appears to be quicker than optimization that occurs naturally (artificial selection vs natural selection), at least for specific vectors and desired outcomes. Would there come a point where genetic algorithms result in a more complex system than a DNA based lifeform? It has been stated by people smarter than me that the only way to simulate how a particular DNA strand would develop into a living creature would be in real-time in a real environment (ie the only way to see how a person would grow is to grow a person). Will this always be true?


I like the idea. I suppose that it wouldn't be as impenetrable to the researchers who created it as biology is to the people studying it now, but I'd tend to expect that by the time we had some system of nested genetic algorithms doing their thing quite that autonomously, we'd probably also separately have a stronger handle on biology, so.

cellocgw wrote:Yeah, but....

How much ya wanna bet nobody ever put comments into the DNA source code? :P


Why continue the source code metaphor? DNA is written compiled.

Exactly so, and evolution is analogous to trying to improve the program by twerking the bits of the executable. It's a pretty dumb way to try to improve a design, and as Dawkins points out in The Greatest Show on Earth, the "design" of living things is anything but intelligent. He talks about this nerve which connects the larynx to the brain, which in all mammals takes a circuitous route down to the chest; in the Giraffe the detour is several metres long. This makes me doubt the general benefit of genetic algorithms. I know they've been used with some success in antenna design; basically I can see the appeal in fields like this where the interaction between the variables is just too complicated to be able to design using reason and mathematics. Still, even in that case, it strikes me that they are really just one example of optimisation using a cost function, and you might do better by setting out to design that process more intelligently. For the moment, nature has done in some ways a better job of "designing" intelligence than we have, but then that's partly true by definition, in that what we think of as intelligence is more-or-less defined by the human variety (this assumption is built right into the Turing Test, for instance).
xtifr wrote:... and orthogon merely sounds undecided.

rmsgrey
Posts: 3616
Joined: Wed Nov 16, 2011 6:35 pm UTC

Re: 1605: "DNA"

Postby rmsgrey » Wed Nov 18, 2015 5:43 pm UTC

Human optimisation may be faster, but the result tends to be less optimised because when humans do it, we still expect it to make sense to us. When evolution optimises something, there's no consideration of whether there's a comprehensible story to be told of how it works or not - it just produces something that works... somehow.

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: 1605: "DNA"

Postby Izawwlgood » Wed Nov 18, 2015 5:45 pm UTC

Genetic algorithms are useful for some contexts but not others. Efficiency, maybe not so much, feature generation or existing feature refinement, maybe better suited.

For example, to use the circuitous vagus nerve - nerve signal propagation is really fast, and there's not much pressure to route it more efficiently if the nerve is already doing it's job. The increased energy cost of maintaining a single nerve bundle is probably quite minor relative to a number of other gross anatomical features giraffes had to adapt to (knees, valves in arteries, stronger vertebrae, etc), so as always with evolution, it isn't about making the best and most efficient, but the thing that works.
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

commodorejohn
Posts: 1176
Joined: Thu Dec 10, 2009 6:21 pm UTC
Location: Placerville, CA
Contact:

Re: 1605: "DNA"

Postby commodorejohn » Wed Nov 18, 2015 5:57 pm UTC

cellocgw wrote:How much ya wanna bet nobody ever put comments into the DNA source code? :P

God is obviously a Real Programmer.
"'Legacy code' often differs from its suggested alternative by actually working and scaling."
- Bjarne Stroustrup
www.commodorejohn.com - in case you were wondering, which you probably weren't.

User avatar
Whizbang
The Best Reporter
Posts: 2238
Joined: Fri Apr 06, 2012 7:50 pm UTC
Location: New Hampshire, USA

Re: 1605: "DNA"

Postby Whizbang » Wed Nov 18, 2015 6:01 pm UTC

I've occasionally wondered if perhaps there is room in the programming world for a code commenter, by which I mean a person is part of the dev process whose duty it is to come along behind the programmers and add comments. Obviously there will need to be continuous communication with the programmers.

User avatar
orthogon
Posts: 3057
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1605: "DNA"

Postby orthogon » Wed Nov 18, 2015 6:02 pm UTC

Izawwlgood wrote:Genetic algorithms are useful for some contexts but not others. Efficiency, maybe not so much, feature generation or existing feature refinement, maybe better suited.

For example, to use the circuitous vagus nerve - nerve signal propagation is really fast, and there's not much pressure to route it more efficiently if the nerve is already doing it's job. The increased energy cost of maintaining a single nerve bundle is probably quite minor relative to a number of other gross anatomical features giraffes had to adapt to (knees, valves in arteries, stronger vertebrae, etc), so as always with evolution, it isn't about making the best and most efficient, but the thing that works.

Sure, and maybe the vagus nerve (thanks for reminding me) is a good example of something that looks ridiculous as opposed to actually being a bad design where the obvious improvement would be beneficial; on the other hand there are lots of examples in that chapter of things that really don't work very well: our quadruped backs that are easily damaged by lifting heavy objects in a standing position, for example. The trouble is that improvements can only be made by a series of tiny steps each of which must be an improvement over (or at least no worse than) the previous "design". Even if rerouting the vagus nerve conferred a significant advantage (and it probably would; after all the benefits from mutations that drive evolution are surely even smaller), it couldn't ever happen, because (as I understand it) it's somehow the wrong side of something else: you'd have to "unplug it" to route it around the other way, and it's not so easy to see how to get something the other side of something else in small steps that result in viable intermediate organisms.
xtifr wrote:... and orthogon merely sounds undecided.

User avatar
Whizbang
The Best Reporter
Posts: 2238
Joined: Fri Apr 06, 2012 7:50 pm UTC
Location: New Hampshire, USA

Re: 1605: "DNA"

Postby Whizbang » Wed Nov 18, 2015 6:09 pm UTC

One case I find more compelling than the recurrent laryngeal nerve is the human eye (just look at it! It must be designed!). This thing is a patch upon a patch upon a kludge that then needs some serious post-processing in the brain to get an acceptable image. Less simple than the nerve example, sure, but really hits home the idea of incremental, non-intelligent improvements.

User avatar
Copper Bezel
Posts: 2426
Joined: Wed Oct 12, 2011 6:35 am UTC
Location: Web exclusive!

Re: 1605: "DNA"

Postby Copper Bezel » Wed Nov 18, 2015 6:12 pm UTC

orthogon wrote:Sure, and maybe the vagus nerve (thanks for reminding me) is a good example of something that looks ridiculous as opposed to actually being a bad design where the obvious improvement would be beneficial; on the other hand there are lots of examples in that chapter of things that really don't work very well: our quadruped backs that are easily damaged by lifting heavy objects in a standing position, for example. The trouble is that improvements can only be made by a series of tiny steps each of which must be an improvement over (or at least no worse than) the previous "design". Even if rerouting the vagus nerve conferred a significant advantage (and it probably would; after all the benefits from mutations that drive evolution are surely even smaller), it couldn't ever happen, because (as I understand it) it's somehow the wrong side of something else: you'd have to "unplug it" to route it around the other way, and it's not so easy to see how to get something the other side of something else in small steps that result in viable intermediate organisms.


If that's the case, it could well be a developmental process constraint. Like the marsupials' ability to take up basically every mammal role except the ungulates, since the infant form must have grasping hands for the rearing strategy, so we end up with giant rabbit beasts instead.

And neither evolution nor genetic algorithms based on the process is strictly step to step if you permit hybridization.

Not to say that there aren't going to be awkward constraints. There always will be, including the culture and habit of engineers on drafting tables.
So much depends upon a red wheel barrow (>= XXII) but it is not going to be installed.

she / her / her

User avatar
Izawwlgood
WINNING
Posts: 18686
Joined: Mon Nov 19, 2007 3:55 pm UTC
Location: There may be lovelier lovelies...

Re: 1605: "DNA"

Postby Izawwlgood » Wed Nov 18, 2015 6:31 pm UTC

Copper Bezel wrote:Like the marsupials' ability to take up basically every mammal role except the ungulates
Kangaroos are pretty ungulate like!
... with gigantic melancholies and gigantic mirth, to tread the jeweled thrones of the Earth under his sandalled feet.

User avatar
orthogon
Posts: 3057
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1605: "DNA"

Postby orthogon » Wed Nov 18, 2015 6:36 pm UTC

Whizbang wrote:One case I find more compelling than the recurrent laryngeal nerve is the human eye (just look at it! It must be designed!). This thing is a patch upon a patch upon a kludge that then needs some serious post-processing in the brain to get an acceptable image. Less simple than the nerve example, sure, but really hits home the idea of incremental, non-intelligent improvements.

Dawkins makes that point too: the clincher being the way some idiot has put the damn retina on backwards!

We engineers are not immune to these very problems ourselves, particularly in software (and probably only because software systems are by far the most complex things that humanity has created). We have bad design that we'd like to sort out, but we'd have to completely refactor, whereas it's quicker and easier to apply a kludge. There's a further analogy with evolution: if we did refactor, some things would stop working and new bugs would be introduced; in other words the modified version might be less "fit" than the original spaghetti code, even though we have the enormous advantage of being able to make a large number of carefully designed and mutually dependent changes in one iteration.
xtifr wrote:... and orthogon merely sounds undecided.

J%r
Posts: 43
Joined: Sun Aug 16, 2015 12:02 pm UTC

Re: 1605: "DNA"

Postby J%r » Wed Nov 18, 2015 7:25 pm UTC

Whizbang wrote:I've occasionally wondered if perhaps there is room in the programming world for a code commenter, by which I mean a person is part of the dev process whose duty it is to come along behind the programmers and add comments. Obviously there will need to be continuous communication with the programmers.


No, code should explain itself, otherwise it isn't clean code. Besides comments are lies! They are often not maintained and are actually about how something used to work, but it works differently now. The only place comments are useful is in an API. And there are ways to test if API comments are up to date and contain everything.

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: 1605: "DNA"

Postby Tyndmyr » Wed Nov 18, 2015 7:35 pm UTC

Whizbang wrote:I've occasionally wondered if perhaps there is room in the programming world for a code commenter, by which I mean a person is part of the dev process whose duty it is to come along behind the programmers and add comments. Obviously there will need to be continuous communication with the programmers.


It's not the worst idea. I'd give it a shot.

It seems way more practical than pair programming, and potentially better because...not being the one who was involved in coding it is really helpful for knowing what comments are essential.

When left to my own devices, I comment almost nothing, relying on structure, convention, and descriptive names unless I've done something particularly "clever".

rmsgrey
Posts: 3616
Joined: Wed Nov 16, 2011 6:35 pm UTC

Re: 1605: "DNA"

Postby rmsgrey » Wed Nov 18, 2015 7:41 pm UTC

Tyndmyr wrote:
Whizbang wrote:I've occasionally wondered if perhaps there is room in the programming world for a code commenter, by which I mean a person is part of the dev process whose duty it is to come along behind the programmers and add comments. Obviously there will need to be continuous communication with the programmers.


It's not the worst idea. I'd give it a shot.

It seems way more practical than pair programming, and potentially better because...not being the one who was involved in coding it is really helpful for knowing what comments are essential.

When left to my own devices, I comment almost nothing, relying on structure, convention, and descriptive names unless I've done something particularly "clever".


There are at least three types of comments:
- TODO/FIXME - comments a programmer leaves for themself to come back to tomorrow or next week or whenever there's a spare moment
- Weirdness alerts - comments a programmer leaves for whoever has to maintain that code in future (possibly themself) calling out something non-obvious or non-standard about the code and its interactions
- Formal documentation - comments a programmer comes back and adds just before a code review in order to satisfy some sort of imposed standard, which may or may not actually be useful to anyone

User avatar
orthogon
Posts: 3057
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1605: "DNA"

Postby orthogon » Wed Nov 18, 2015 7:49 pm UTC

J%r wrote:
Whizbang wrote:I've occasionally wondered if perhaps there is room in the programming world for a code commenter, by which I mean a person is part of the dev process whose duty it is to come along behind the programmers and add comments. Obviously there will need to be continuous communication with the programmers.


No, code should explain itself, otherwise it isn't clean code. Besides comments are lies! They are often not maintained and are actually about how something used to work, but it works differently now. The only place comments are useful is in an API. And there are ways to test if API comments are up to date and contain everything.

I think this is right, generally. One of my most hated things are those huge comment blocks before a function that list the arguments that it used to have and what they used to mean. Describing in more detail what the arguments do should be done right next to the declaration of that argument.

What I would like to see is not a code commenter, but a code commentator, in the style of a sports commentator. "Smith passes an object ... Jones calls a method ... he commits ... and it's a build! And just look at that scoreline: no errors, no warnings!" Of course, the programmers would have to jump on each other and embrace, which may be an issue for the less socially comfortable of us.
xtifr wrote:... and orthogon merely sounds undecided.

CharonPDX
Posts: 60
Joined: Wed Apr 27, 2011 4:55 am UTC

Re: 1605: "DNA"

Postby CharonPDX » Wed Nov 18, 2015 7:57 pm UTC

HES wrote:At 213 thousand characters, it's about four times over the post limit. I'm sure people are capable of the three or so clicks required to see it for themselves.


For such a simple-looking web page, that is insanely much code...

User avatar
rhomboidal
Posts: 796
Joined: Wed Jun 15, 2011 5:25 pm UTC
Contact:

Re: 1605: "DNA"

Postby rhomboidal » Wed Nov 18, 2015 8:19 pm UTC

We will never, ever unravel the genetic code for Google devs.

User avatar
Flumble
Yes Man
Posts: 2227
Joined: Sun Aug 05, 2012 9:35 pm UTC

Re: 1605: "DNA"

Postby Flumble » Wed Nov 18, 2015 8:26 pm UTC

CharonPDX wrote:
HES wrote:At 213 thousand characters, it's about four times over the post limit. I'm sure people are capable of the three or so clicks required to see it for themselves.


For such a simple-looking web page, that is insanely much code...

The code likely contains:
- page layout
- interactive results
- GNU Terry Pratchett
- Love's Labour's Won
- The answer to life, the universe and everything
- A COBOL Go interpreter to Javascript compiler in Javascript a bootstrapped compiler written in Go
- NSA backdoor
- Bitcoin miner
- That time-travel code Bender had in that one episode
- Fake bitcoin miner that's really just trying to crack Starbuck's wifi password so the guys at google have free protected wifi during a coffee break
- All future doodles, so you see them even if you're in offline mode

That's insanely good compression for such a load of code.

GuesssWho
Posts: 19
Joined: Thu Jan 29, 2015 9:29 am UTC

Re: 1605: "DNA"

Postby GuesssWho » Wed Nov 18, 2015 11:32 pm UTC

This is why I find people who trust GMOs to be sad. 'But it's just like selective breeding,' they whine.

No, no it is not. For the same reason that a cabinet is not a bonsai.

Lets see what happens when you semi-randomly take code from one website and shove it into the coding of another website. Probably nothing helpful.

Plus, we've nearly killed the world with it once already that I know of . . .

commodorejohn
Posts: 1176
Joined: Thu Dec 10, 2009 6:21 pm UTC
Location: Placerville, CA
Contact:

Re: 1605: "DNA"

Postby commodorejohn » Thu Nov 19, 2015 12:01 am UTC

GuesssWho wrote:Lets see what happens when you semi-randomly take code from one website and shove it into the coding of another website. Probably nothing helpful.

That's like half of what passes for professional web design these days.
"'Legacy code' often differs from its suggested alternative by actually working and scaling."
- Bjarne Stroustrup
www.commodorejohn.com - in case you were wondering, which you probably weren't.


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: No registered users and 80 guests