Programming isn't hard...
-
I just went for
item. And yes, it is to remain abstract. It is an Interface Method that has a T (as in generic) as a parameter. That is the reason I did not want to go forobj. Because the parameter is of type T, which is more specific than an Object.It's an OO world.
public class Naerling : Lazy<Person>{}
void DoSomethingGeneric(T t){ ... } If it's just any T, with no further restrictions, the fact it is a 'T' is the only thing that identifies it, so I'd call it 't'. The fact that it's a T that something is done with should be clear from the method name.
-
This is a particularly stupid tip and would never work. Apart from the fact that it would require superhuman mental gymnastics, once you have more than 42 variables, in order to maintain alphabetic ordering, it is yet another invidious ploy by the militant fundamentalist wing of the ASL (Anti-Southpaw League). Please remember that The Lounge is no place for these religious outpourings.
Henry Minute Do not read medical books! You could die of a misprint. - Mark Twain Girl: (staring) "Why do you need an icy cucumber?" “I want to report a fraud. The government is lying to us all.” I wouldn't let CG touch my Abacus! When you're wrestling a gorilla, you don't stop when you're tired, you stop when the gorilla is.
I agree. Since all the syntactic characters are on the other side of the keyboard (parens, braces, semicolon, equals and single quote; double quote too for Americans), clearly variable names ought to use exclusively the right side of the keyboard (p0, p9, p8, ... p5, o0, ... etc).
-
Thinking of names for your classes and variables, THAT's hard... :doh: Frustration #1 of the evening... :sigh:
It's an OO world.
public class Naerling : Lazy<Person>{}
it's easy when you coding some useless crap, real programming never was easy and will never be :~
d{^__^}b - it's time to fly
-
it's easy when you coding some useless crap, real programming never was easy and will never be :~
d{^__^}b - it's time to fly
It's called irony :)
It's an OO world.
public class Naerling : Lazy<Person>{}
-
When Black and Scholes were figuring out how to value options, though, they probably were using one- and two-letter variable names, and they ended up winning the Nobel Prize. Maybe the difference is that people don't have to "maintain" mathematics. There was a movie where Leo DiCaprio (or maybe Matt Damon?) was a maintenance man who did mathematics... but that's a different story. :-D
I always get those two mixed up, made watching The Departed very difficult to begin with :)
-
Dijkstra didn't have to debug an option valuation model written in C++ by a math PhD... True story... Almost every variable in the Black-Scholes calculation was one or two letters long, and none of them had any relevance to what they represented. Don't get me wrong... It (mostly) worked, and it was fast, but... Ow, my eyes!
Proud to have finally moved to the A-Ark. Which one are you in?
Author of the Guardians Saga (Sci-Fi/Fantasy novels)I know what you mean - my colleague (who writes the B-S module for his option pricing program (which I maintain)) loves names like henry, peter, peterp, peter2 etc - I'm just glad I don't have to work on that bit of code. I used to employ a programmer who was very good, and very hard working, but he never used any kind of meaningful names for anything, so a complex C function representing, say, a vector drawing interpreter, would contain tens of variables named a,aa,aaa,aaa, x,xx,xxx,xxxx, i,ii,iii,iiii,iiiii and so forth... 8)
-
Ian Shlasko wrote:
Tenagra t = new Tenagra(); t.Add(new Darmok()); t.Add(new Jilad()); t.Add(new TheBeast()); t.EpicBattle(); t.MoveItemsTo(new Ocean());
Congratulations, this is the nerdiest thing I have read this month, on many levels. And considering the crap I read, that was an achievement.
Temba(his)::arms(wide){} ;P
-
This is Dijkstra's opinion, not mine! But I think he would tell you that mathematicians get by with X, Y, theta, pi, etc., and that programming is just a species of mathematics. After all, Einstein didn't discover that realEnergy = realMass * [ (squareRootOfTheSpeedOfLight)² ] Rather, he was just fine with E=mc², and (amazingly, if all you know is programming) people still embraced his findings.
_beauw_ wrote:
After all, Einstein didn't discover that
realEnergy = realMass * [ (squareRootOfTheSpeedOfLight)² ]He didn't.
realEnergy = realMass * [ (theSpeedOfLightInVacuum)² ]
FTFY
-
Thinking of names for your classes and variables, THAT's hard... :doh: Frustration #1 of the evening... :sigh:
It's an OO world.
public class Naerling : Lazy<Person>{}
Back in College in 1995, a friend was working on his first RPG program. "What should I call this Dumaflopper?" He asked out loud. I said, "You know, you should just call every program this semester Dumaflopper." He did. Every assignment he turned in was dumaflopper.rpg.
-
OriginalGriff wrote:
Star Fleet don't seem to have any truck with H&S legislation, or they wouldn't fill all the computer consoles with semtex...
Yeah, I always wondered why it was that they could manage matter-energy conversion, subspace communication, and warp drive... But couldn't seem to figure out how to make a surge protector or... *gasp*... a FUSE!
Proud to have finally moved to the A-Ark. Which one are you in?
Author of the Guardians Saga (Sci-Fi/Fantasy novels)i always wondered that myself, its like they shunted warp plasma to run a frikken computer screen, and dont have anything to manage power surges. i think the poor ensigns on the terminals would rather have to switch a breaker than have their console explode in their face and kill them. while on the subject: seatbelts. how many poor ensigns and redshirts have met their fate from being thrown across the bridge and breaking their neck. seatbelts save lives, or at least save you from having to go back to a star base to get more troops. "darmok and jalad at tanagra" "i dont understand" "DARMOK AND JALAD AT TANAGRA!" great episode.
-
This is Dijkstra's opinion, not mine! But I think he would tell you that mathematicians get by with X, Y, theta, pi, etc., and that programming is just a species of mathematics. After all, Einstein didn't discover that realEnergy = realMass * [ (squareRootOfTheSpeedOfLight)² ] Rather, he was just fine with E=mc², and (amazingly, if all you know is programming) people still embraced his findings.
Yes, but the reason he got away with E = mc2 is because physicists had, over the course many years, established conventions for the use of the symbols. A programmer must establish conventions in application domains that they are making up as they are going along. Ouch!
-
Yeah, it's true. The theme of many of Dijkstra's arguments was that natural language is a poor vehicle for discussing problems in computation, and his writing style seems at times to be intended to prove this assertion! I have read through his "Meaningful Identifiers" paper, though, and in summary, his basic point seems to be that attempting to give variables descriptive natural language names is a trap. A natural language name can never truly express what a variable is, fundamentally, or what it does, only provide hints. Because these natural language names are subjective and ambiguous, a danger occurs when people (e.g. maintenance programmers) attempt to extract meaning from them instead of just reading the code, e.g.
#define max_subscript 4
//...
int some_array[max_subscript];
The declaration at the end of this snippet implies that the name
max_subscriptis actually an inaccurate description. The real maximum subscript is 3. But in order to detect this, one must 1) be familiar with C syntax and 2) actually see and consider this declaration (versus simply seeing the incorrectly named constant used elsewhere). As a result, we would be better off ifmax_subscriptwere simply calledM(oriLikeFishTacos,brett_favre, etc.) Dijkstra's argument seems sound, but in practice I do give my identifiers meaningful natural-language names, because people expect this. It's not difficult to extend Dijkstra's argument about identifier names to comments as well... consider the following:void a_function()
{
int * a = new int; //Allocate "a" on the stack//...
}//All that stuff we allocated on the stack gets cleaned up automatically here
Again, this is a simplistic example (that
intisn't on the stack and it won't get cleaned up automatically). Hopefully, though, the message is clear: we'd actually be better off without those supposedly helpful comments. As an experienced programmer, I have suffered through a few scenarios where an inaccurate comment actually did slow me down. However, I must be careful to point out that none of these extreme positions reflect my own personal programming practice... (although I do think that developers should at least acquaint themselves with Dijkstra's argument)._beauw_ wrote:
.. his basic point seems to be that attempting to give variables descriptive natural language names is a trap. A natural language name can never truly express what a variable is, fundamentally, or what it does, only provide hints.
Sounds to me like Dijkstra completely missed the point of giving natural language names to variables. It isn't to try to describe the variable and its every nuance, it's to provide a mnemonic word or phrase that triggers the memory in the programmer's head of what what the variable holds and its every nuance. For a few, single letters suffice, for the rest of us, a word or phrase works better. Einstein's E=mc2 is a perfect example of mnemonics. "E" stands for energy, "m" for mass and "c" is the standard mnemonic letter used to represent the speed of light. His variable choice wasn't random, it was mnemonic. Pretty amazing it actually works as a mnemonic for us non-mathematician/non-physicist types too. The max_subscript example is bad because the mnemonic itself triggers a misunderstanding, which then fights with the true understanding.
We can program with only 1's, but if all you've got are zeros, you've got nothing.
-
Back in College in 1995, a friend was working on his first RPG program. "What should I call this Dumaflopper?" He asked out loud. I said, "You know, you should just call every program this semester Dumaflopper." He did. Every assignment he turned in was dumaflopper.rpg.
Where they flops too? :laugh:
It's an OO world.
public class Naerling : Lazy<Person>{}
-
I prefer to call all my variables a1, a2, a3, a4, and so on. Also, once I get to a6, I use another letter (say, b1)... that way, I only ever have to use my left hand and can keep my right hand on the mouse. It's a very efficient technique I wish everybody else would adopt. Until then, I just refactor all the code I see with overly long variable names like "count" (that's a particularly bad name because some of the characters require the right hand to type). I sleep easily at night, comfortable in the knowledge that all my refactoring has made the world a better place. :-\
Somebody in an online forum wrote:
INTJs never really joke. They make a point. The joke is just a gift wrapper.
Saving the right hand for better uses?
-
Where they flops too? :laugh:
It's an OO world.
public class Naerling : Lazy<Person>{}
Naerling wrote:
Where they flops too?
Wow, he's going for the record in grammatical errors in 4 words ;-)
-
Naerling wrote:
Where they flops too?
Wow, he's going for the record in grammatical errors in 4 words ;-)
Were* And that's the only one I can find... But let's see how well your Dutch grammar is :) Als je durft ;p (yeah, translate that!)
It's an OO world.
public class Naerling : Lazy<Person>{}
-
_beauw_ wrote:
The declaration at the end of this snippet implies that the name
max_subscriptis actually an inaccurate description.I would agree, but that isn't a fault of the naming convention. Rather, it is a fault of the coder: max_subscript is wrong, and doesn't describe what the constant is used for: max_elements_count would have been better (or max_whatevertheheckthearrayholds_count even better still) For me, the problem is I grew up with Fortran, where a line of code was 80 characters, and variables were a max of six characters. Generally names were impenetrable, short and unrelated to the use. So, when I can I use "proper" names for things, because it means I can read the code in six months time and stand a vague chance of working out how it works. :laugh:
Real men don't use instructions. They are only the manufacturers opinion on how to put the thing together. Manfred R. Bihy: "Looks as if OP is learning resistant."
When I learnt Algol60 (no limit on variable names) after several years of writing FORTRAN IV / 66 (max 6 chars in variable names), our lecturer said we had to use long, meaningful identifiers and I failed to see what the advantage was. So, for an assessed piece of work, I used 80-char identifiers (as identifiers could not be broken across cards) causing even simple assignments to straddle multiple cards and to render the whole program totally unintelligible. [Using whole cards per identifier did make coding easy - I could just duplicate the cards rather than having to retype the identifier wherever it was used] I also submitted a legible version using short identifiers. The lecturer was not impressed.
-
When I learnt Algol60 (no limit on variable names) after several years of writing FORTRAN IV / 66 (max 6 chars in variable names), our lecturer said we had to use long, meaningful identifiers and I failed to see what the advantage was. So, for an assessed piece of work, I used 80-char identifiers (as identifiers could not be broken across cards) causing even simple assignments to straddle multiple cards and to render the whole program totally unintelligible. [Using whole cards per identifier did make coding easy - I could just duplicate the cards rather than having to retype the identifier wherever it was used] I also submitted a legible version using short identifiers. The lecturer was not impressed.
:laugh: I can't imagine why not!
Real men don't use instructions. They are only the manufacturers opinion on how to put the thing together. Manfred R. Bihy: "Looks as if OP is learning resistant."
-
_beauw_ wrote:
The declaration at the end of this snippet implies that the name
max_subscriptis actually an inaccurate description.I would agree, but that isn't a fault of the naming convention. Rather, it is a fault of the coder: max_subscript is wrong, and doesn't describe what the constant is used for: max_elements_count would have been better (or max_whatevertheheckthearrayholds_count even better still) For me, the problem is I grew up with Fortran, where a line of code was 80 characters, and variables were a max of six characters. Generally names were impenetrable, short and unrelated to the use. So, when I can I use "proper" names for things, because it means I can read the code in six months time and stand a vague chance of working out how it works. :laugh:
Real men don't use instructions. They are only the manufacturers opinion on how to put the thing together. Manfred R. Bihy: "Looks as if OP is learning resistant."
Reminds me of my SAP days, relatively recently, where table identifiers were normally four character acronyms, and field names six characters , both often based on the original German names. After a surprisingly short while I found these very easy to remember and use, considering we coded in a line editor without Intellisense.
