Code Complete, Second Edition
-
Assuming you're not being sarcastic (I'm assumimg you are being sarcastic however), this seems a little inconsistent with your philosophy: to be advocating English language programming but rejecting descriptively named variables is a contradiction to say the least. Assuming you are being sarcastic, I don’t think your point is valid. The equation “a=pi*r^2” is just as rigid as if “area=pi*radius^2” was used. The names of the variables is not connected to the rigidity of the language rules. Both formality and descriptiveness have their place.
Steve
Stephen Hewitt wrote:
Assuming you're not being sarcastic (which I admit I’m having trouble discerning, I'm leaning towards sarcastic however), this seems a little inconsistent with your philosophy: to be advocating English language programming but rejecting descriptively named variables is a contradiction to say the least.
Well, it seems to me that someone who prefers "isBookGood" to "itbg_TF" should also prefer "If the book is good" to "if (itbg_TF)" or even "if (isBookGood)". Clearly, "isBookGood" is closer to the thought in the programmer's mind than "itbg_TF" which is less readable and drags in all sorts of implementation issues. But isn't "If the book is good" closer to what the programmer is thinking than "if (itbg_TF)" or "if (isBookGood)"?
-
Stephen Hewitt wrote:
Assuming you're not being sarcastic (which I admit I’m having trouble discerning, I'm leaning towards sarcastic however), this seems a little inconsistent with your philosophy: to be advocating English language programming but rejecting descriptively named variables is a contradiction to say the least.
Well, it seems to me that someone who prefers "isBookGood" to "itbg_TF" should also prefer "If the book is good" to "if (itbg_TF)" or even "if (isBookGood)". Clearly, "isBookGood" is closer to the thought in the programmer's mind than "itbg_TF" which is less readable and drags in all sorts of implementation issues. But isn't "If the book is good" closer to what the programmer is thinking than "if (itbg_TF)" or "if (isBookGood)"?
I've edited and added to my OP since you wrote this reply.
Steve
-
Stephen Hewitt wrote:
Assuming you're not being sarcastic (which I admit I’m having trouble discerning, I'm leaning towards sarcastic however), this seems a little inconsistent with your philosophy: to be advocating English language programming but rejecting descriptively named variables is a contradiction to say the least.
Well, it seems to me that someone who prefers "isBookGood" to "itbg_TF" should also prefer "If the book is good" to "if (itbg_TF)" or even "if (isBookGood)". Clearly, "isBookGood" is closer to the thought in the programmer's mind than "itbg_TF" which is less readable and drags in all sorts of implementation issues. But isn't "If the book is good" closer to what the programmer is thinking than "if (itbg_TF)" or "if (isBookGood)"?
The Grand Negus wrote:
But isn't "If the book is good" closer to what the programmer is thinking than "if (itbg_TF)" or "if (isBookGood)"?
Again trying to plug "Plain English Programming" :zzz:
-
I'm thinking about buying the book Code Complete, Second Edition. How many of you have actually read it and how good was it. I know there are things like naming your booleans descriptively like isThisBookGood or buyThisBook instead of itbg_TF. Is there more to this book than that? I want to really learn something that will help me be a better architect of my software and a better coder also. What do you think? Is The Pragmatic Programmer: From Journeyman to Master good also?
█▒▒▒▒▒██▒█▒██ █▒█████▒▒▒▒▒█ █▒██████▒█▒██ █▒█████▒▒▒▒▒█ █▒▒▒▒▒██▒█▒██
Captain See Sharp wrote:
I want to really learn something that will help me be a better architect of my software and a better coder also. What do you think?
Try this little book: "Programming Pearls" by Jon Louis Bentley. But do as he says, not as he does. You might also try "Project Oberon - The Design of an Operating System and Compiler" by N. Wirth and J. Gutknecht which is available free as a PDF here[^]. And, of course, you can always... well, you know.
-
Captain See Sharp wrote:
I want to really learn something that will help me be a better architect of my software and a better coder also. What do you think?
Try this little book: "Programming Pearls" by Jon Louis Bentley. But do as he says, not as he does. You might also try "Project Oberon - The Design of an Operating System and Compiler" by N. Wirth and J. Gutknecht which is available free as a PDF here[^]. And, of course, you can always... well, you know.
The Grand Negus wrote:
And, of course, you can always... well, you know.
Is that your signature?:rolleyes:
-
I've edited and added to my OP since you wrote this reply.
Steve
Stephen Hewitt wrote:
The names of the variables is not connected to the rigidity of the language rules. Both formality and descriptiveness have their place.
Agreed. But we shouldn't have to be more formal than necessary, or less descriptive than we can be - ever. And if our goal is to produce an artificial intelligence, we will probably find it easier to work in a language that is about as formal and about as rigid as the machine we're trying to program - no more and no less. But why limit ourselves? This statement, for example, is perfectly reasonable in a natural language framework: If the book is good, put sqrt(sqr(length)+sqr(width)) into the book's diagonal. Here, the thought is expressed in natural terms except where a more specialized syntax makes the job "easier". In other words, it's a trivial matter to include rigid, formal syntax within a natural language framework; the reverse is much more difficult to accomplish.
-
The Grand Negus wrote:
But isn't "If the book is good" closer to what the programmer is thinking than "if (itbg_TF)" or "if (isBookGood)"?
Again trying to plug "Plain English Programming" :zzz:
Pravarakhya wrote:
Again trying to plug "Plain English Programming"
Actually, I'm asking what makes "isBookGood" better than "itbg_TF", and whether or not those same virtues can be used to evaluate procedural code as well as data definition code. If the virtues in question can be delineated and apply in both circumstances, that would appear to be a point in favor of a natural language approach to programming; otherwise, it would be a point against it.
-
I'm thinking about buying the book Code Complete, Second Edition. How many of you have actually read it and how good was it. I know there are things like naming your booleans descriptively like isThisBookGood or buyThisBook instead of itbg_TF. Is there more to this book than that? I want to really learn something that will help me be a better architect of my software and a better coder also. What do you think? Is The Pragmatic Programmer: From Journeyman to Master good also?
█▒▒▒▒▒██▒█▒██ █▒█████▒▒▒▒▒█ █▒██████▒█▒██ █▒█████▒▒▒▒▒█ █▒▒▒▒▒██▒█▒██
Captain See Sharp wrote:
I'm thinking about buying the book Code Complete, Second Edition.
So you should, it's come up a lot here of late.
Captain See Sharp wrote:
How many of you have actually read it
A lot, I'd hope.
Captain See Sharp wrote:
how good was it
I wish I'd read it sooner. As it was, it verbalised a lot of stuff I already was doing, but was still an excellent read. If I ever hire anyone, I'll buy each new hire a copy.
Captain See Sharp wrote:
I know there are things like naming your booleans descriptively like isThisBookGood or buyThisBook instead of itbg_TF. Is there more to this book than that
A lot more. The bit on writing short methods and refactoring is worth the price of admission alone.
Captain See Sharp wrote:
I want to really learn something that will help me be a better architect of my software and a better coder also. What do you think?
Buy it. The worst that can happen is what happened to me, you'll be forced to think about things you've just been doing intuitively ( I'm not suggesting I learned nothing, just that a lot of stuff I read, I had learned from experience, but often not really thought about so much )
Captain See Sharp wrote:
Is The Pragmatic Programmer: From Journeyman to Master good also?
I've not read it, it could be.
Christian Graus - Microsoft MVP - C++ Metal Musings - Rex and my new metal blog
-
Stephen Hewitt wrote:
Assuming you're not being sarcastic (which I admit I’m having trouble discerning, I'm leaning towards sarcastic however), this seems a little inconsistent with your philosophy: to be advocating English language programming but rejecting descriptively named variables is a contradiction to say the least.
Well, it seems to me that someone who prefers "isBookGood" to "itbg_TF" should also prefer "If the book is good" to "if (itbg_TF)" or even "if (isBookGood)". Clearly, "isBookGood" is closer to the thought in the programmer's mind than "itbg_TF" which is less readable and drags in all sorts of implementation issues. But isn't "If the book is good" closer to what the programmer is thinking than "if (itbg_TF)" or "if (isBookGood)"?
The Grand Negus wrote:
But isn't "If the book is good" closer to what the programmer is thinking than "if (itbg_TF)" or "if (isBookGood)"?
Only if the programmer doesn't know any language apart from English. If a programming language existed that was able to understand plain english, and get over the hurdle that you have to either simply provide a new syntax as rigid as existing languages ( which PE does ), or find a way to take all manner of statements and yet work out the programmers precise intentions, then that would be a clearer way to go for the lay person. As it stands, PE just defines it's own rigid syntax, and no language is easier than another, they are just different ( I find it a pain to read over verbose code, like VB and PE )
Christian Graus - Microsoft MVP - C++ Metal Musings - Rex and my new metal blog
-
I detest it. I found it very pompous and dogmatic with the few good concepts hidden amongst some really bad ideas and generalizations. As I've said in past critiques of the book, I still say read it through with a skeptical eye then toss it in the trash.
Anyone who thinks he has a better idea of what's good for people than people do is a swine. - P.J. O'Rourke
Joe Woodbury wrote:
I found it very pompous and dogmatic
Don't hold back, tell us how you really fell. :)
Joe Woodbury wrote:
As I've said in past critiques of the book,
Link? I would be interested in your take on it, I believe yours is about the only voice of dissension when it comes to Code Complete.
I'd love to help, but unfortunatley I have prior commitments monitoring the length of my grass. :Andrew Bleakley:
-
The Grand Negus wrote:
But isn't "If the book is good" closer to what the programmer is thinking than "if (itbg_TF)" or "if (isBookGood)"?
Only if the programmer doesn't know any language apart from English. If a programming language existed that was able to understand plain english, and get over the hurdle that you have to either simply provide a new syntax as rigid as existing languages ( which PE does ), or find a way to take all manner of statements and yet work out the programmers precise intentions, then that would be a clearer way to go for the lay person. As it stands, PE just defines it's own rigid syntax, and no language is easier than another, they are just different ( I find it a pain to read over verbose code, like VB and PE )
Christian Graus - Microsoft MVP - C++ Metal Musings - Rex and my new metal blog
Christian Graus wrote:
If a programming language existed that was able to understand plain english, and get over the hurdle that you have to either simply provide a new syntax as rigid as existing languages ( which PE does ), or find a way to take all manner of statements and yet work out the programmers precise intentions, then that would be a clearer way to go for the lay person.
First of all, you're mistaken when you suggest that the syntax of Plain English is as rigid as other existing languages. It's not. I freely admit that it's not as flexible as we'd like it to be, but it's getting there. Secondly, we agree that the ultimate goal is to "take all manner of statements and yet work out the programmers [or layman's] precise intentions". And we've got a succinct, executable plan to make that happen within the next few years.
-
The Grand Negus wrote:
But isn't "If the book is good" closer to what the programmer is thinking than "if (itbg_TF)" or "if (isBookGood)"?
Only if the programmer doesn't know any language apart from English. If a programming language existed that was able to understand plain english, and get over the hurdle that you have to either simply provide a new syntax as rigid as existing languages ( which PE does ), or find a way to take all manner of statements and yet work out the programmers precise intentions, then that would be a clearer way to go for the lay person. As it stands, PE just defines it's own rigid syntax, and no language is easier than another, they are just different ( I find it a pain to read over verbose code, like VB and PE )
Christian Graus - Microsoft MVP - C++ Metal Musings - Rex and my new metal blog
Christian Graus wrote:
or find a way to take all manner of statements and yet work out the programmers precise intentions, then that would be a clearer way to go for the lay person.
That would be "true" lexical Analysis, which I have said would be a far better technique than his syntax method (or advanced Zork compiler). Lexical Analysis is an interesting method and has some great work out of several large universities in the AI realm... but then he said all of the AI world is going the wrong direction.... I did my first lexical Analysis work right out of high school, LANA was able to recognize more than his compiler, and actually learned. It was still syntax based, but anything it didn't recognize it would ask about and learn by replacing nouns and verbs. You could say "compare x equal to y" and if it only recognized "if x equal to y" then it would prompt you for a synonym of compare and add it to the dictionary. It had some more advanced work in recognizing noun verb object structure through various types of sentence structures. But all in all it was a pretty nifty lexical analysis work for an 18 year old, replaced my ProDOS and AppleDOS Operating system with something more entertaining, and even looked pretty (think 18 year old interests :rolleyes: and the name LANA :rolleyes: ).
_________________________ Asu no koto o ieba, tenjo de nezumi ga warau. Talk about things of tomorrow and the mice in the ceiling laugh. (Japanese Proverb)
-
Christian Graus wrote:
If a programming language existed that was able to understand plain english, and get over the hurdle that you have to either simply provide a new syntax as rigid as existing languages ( which PE does ), or find a way to take all manner of statements and yet work out the programmers precise intentions, then that would be a clearer way to go for the lay person.
First of all, you're mistaken when you suggest that the syntax of Plain English is as rigid as other existing languages. It's not. I freely admit that it's not as flexible as we'd like it to be, but it's getting there. Secondly, we agree that the ultimate goal is to "take all manner of statements and yet work out the programmers [or layman's] precise intentions". And we've got a succinct, executable plan to make that happen within the next few years.
The Grand Negus wrote:
And we've got a succinct, executable plan to make that happen within the next few years.
I wish you the best of luck. I'd sure love to play with it, when it becomes that good. I predict a middle period where you can't predict what it will do if someone other than you is programming it, tho.
Christian Graus - C++ MVP 'Why don't we jump on a fad that hasn't already been widely discredited ?' - Dilbert
-
I'm thinking about buying the book Code Complete, Second Edition. How many of you have actually read it and how good was it. I know there are things like naming your booleans descriptively like isThisBookGood or buyThisBook instead of itbg_TF. Is there more to this book than that? I want to really learn something that will help me be a better architect of my software and a better coder also. What do you think? Is The Pragmatic Programmer: From Journeyman to Master good also?
█▒▒▒▒▒██▒█▒██ █▒█████▒▒▒▒▒█ █▒██████▒█▒██ █▒█████▒▒▒▒▒█ █▒▒▒▒▒██▒█▒██
yesterday I started reading it and it seemed great. By the way in the first 98 pages I saw that all that was told was obvious, but very interesting... like a situation in which you know everything but it helps you to listen to it again. I think that it can improve the way in which all of us "program"... (and I use "" because it seems that programming is not what we think at the beginning). ;) Hope this helps.
-
I'm thinking about buying the book Code Complete, Second Edition. How many of you have actually read it and how good was it. I know there are things like naming your booleans descriptively like isThisBookGood or buyThisBook instead of itbg_TF. Is there more to this book than that? I want to really learn something that will help me be a better architect of my software and a better coder also. What do you think? Is The Pragmatic Programmer: From Journeyman to Master good also?
█▒▒▒▒▒██▒█▒██ █▒█████▒▒▒▒▒█ █▒██████▒█▒██ █▒█████▒▒▒▒▒█ █▒▒▒▒▒██▒█▒██
Captain See Sharp wrote:
better architect of my software and a better coder also
Whilst I found both editions of CC good, I would also recommend Mike Gunderloy's Coder to Developer[^]. I found that this had a lot more in terms of practical ideas, especially for small teams / solo programmers. Joel "on-software[^]" Spolsky wrote the forward (and you can do worse than read his articles on sw development - they are available either on his site in article format or you can buy a selection in book format from amazon). That said, you won't be dissapointed with CC.
ChrisB ChrisDoesDev[^]
-
I'm thinking about buying the book Code Complete, Second Edition. How many of you have actually read it and how good was it. I know there are things like naming your booleans descriptively like isThisBookGood or buyThisBook instead of itbg_TF. Is there more to this book than that? I want to really learn something that will help me be a better architect of my software and a better coder also. What do you think? Is The Pragmatic Programmer: From Journeyman to Master good also?
█▒▒▒▒▒██▒█▒██ █▒█████▒▒▒▒▒█ █▒██████▒█▒██ █▒█████▒▒▒▒▒█ █▒▒▒▒▒██▒█▒██
Haven't a chance to peek into CC, but PP is marvellous.
-
I'm thinking about buying the book Code Complete, Second Edition. How many of you have actually read it and how good was it. I know there are things like naming your booleans descriptively like isThisBookGood or buyThisBook instead of itbg_TF. Is there more to this book than that? I want to really learn something that will help me be a better architect of my software and a better coder also. What do you think? Is The Pragmatic Programmer: From Journeyman to Master good also?
█▒▒▒▒▒██▒█▒██ █▒█████▒▒▒▒▒█ █▒██████▒█▒██ █▒█████▒▒▒▒▒█ █▒▒▒▒▒██▒█▒██
It is as with sex documentation - "...and if it's good, it's really good!". And it is good. On the surface, much of the information may sound "I knew that" for a seasoned developer. e."The power of variable names" chapter is over 30 pages, packed with thoughts and information. Again, much of it is "I knew that" after you read it (see Douglas Adams, invention of the cat flap). What makes this book a valuable tool is the condensed research results. There are so many "It is better to"'s, but how do you back them up? Sample Fact: Variable names shorter than 8 or longer than 20 increase the debugging effort notably. 10..16 is ideal. (Gorla, Bender, Belander, 1990) This makes it a valuable tool for the project manager, too. And there are the surprises. Things you never thought it would be that way. Do you know that developers are more likely to make mistakes in a small change, as compared to a medium? Did you know that not-so-fast-but-not-so-slow-either developers may well be the worst? Get it! Get it! -- modified at 5:41 Wednesday 10th January, 2007: note to self if you want to be phunny, check your spelling first
Developers, Developers, Developers, Developers, Developers, Developers, Velopers, Develprs, Developers!
We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP
Linkify!|Fold With Us! -
Captain See Sharp wrote:
I know there are things like naming your booleans descriptively like isThisBookGood... instead of itbg_TF.
Why on earth would anyone want to name a boolean "isThisBookGood" instead of "itbg_TF"? The next thing you know they'll be wanting to say things like "If the book is good..." - spaces and all!
Gorla, Benander and benander: Names longer than 20 characters are as bad as shorter than 8.
Developers, Developers, Developers, Developers, Developers, Developers, Velopers, Develprs, Developers!
We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP
Linkify!|Fold With Us! -
Gorla, Benander and benander: Names longer than 20 characters are as bad as shorter than 8.
Developers, Developers, Developers, Developers, Developers, Developers, Velopers, Develprs, Developers!
We are a big screwed up dysfunctional psychotic happy family - some more screwed up, others more happy, but everybody's psychotic joint venture definition of CP
Linkify!|Fold With Us!peterchen wrote:
Names longer than 20 characters are as bad as shorter than 8.
What a curious observation! I take it "Peter" (five characters) isn't working out for you then? And "United States of America" (twenty-four characters) will probably not catch on either; but that's okay, we can just say USA instead - whoops! too short! In any case, the study by Gorla, Benander and Benander that (I believe) you cite is difficult to apply to a language like Plain English because they didn't even consider names with spaces in them. Whichmakesalotofdifference, don't you think?
-
The Grand Negus wrote:
But isn't "If the book is good" closer to what the programmer is thinking than "if (itbg_TF)" or "if (isBookGood)"?
Again trying to plug "Plain English Programming" :zzz: