Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Skyrunner]

You can't have a 3rd option, and still use a boolean.

True, false, and null.  Which would be... kinda appropriate, provided your table design allows it.

Actually that is very language specific. Nullable booleans are a feature of more higher level languages typical these days, but there was a point in the past where they didn't real. Your boolean is actually a pointer to a boolean, and that boolean may only be true of false, but the pointer may be null.

Why, back in my day...

* Skyrunner shakes fist

(More seriously, languages like C, C++, and I think Java don't have null as an option for booleans. Don't quote me on the Java one though.)

[wierd]

In some situations, picking a minimalist datatype and abusing the shit out of it is still somewhat sensible.

Take for instance, an embedded system with mere kilobytes of memory-- OR-- a very very massive database with millions of records inside it that has to all be in memory simultaneously for performance reasons. Having a large structure that maximizes extensibility and flexibility of the object makes sense for situations where the code does not execute continually, or is unlikely to cause IO bounding issues.  For situations where reducing the number of bits being parsed can result in orders of magnitude performance increases, or where the number of bits is simply a limited commodity, abusing minimalist datatypes still makes sense.

[da_nang]

You can't have a 3rd option, and still use a boolean.

True, false, and null.  Which would be... kinda appropriate, provided your table design allows it.

Actually that is very language specific. Nullable booleans are a feature of more higher level languages typical these days, but there was a point in the past where they didn't real. Your boolean is actually a pointer to a boolean, and that boolean may only be true of false, but the pointer may be null.

Why, back in my day...

* Skyrunner shakes fist

(More seriously, languages like C, C++, and I think Java don't have null as an option for booleans. Don't quote me on the Java one though.)

Pretty sure Java's boolean primitive can't have null as a value. Java has the Boolean wrapper class, though, which can.

[Reelya]

(More seriously, languages like C, C++, and I think Java don't have null as an option for booleans. Don't quote me on the Java one though.)

C and C++ allow pointers, which can be Null. C++ and Java allow references. uninitialized (null) pointers or references basically mean you aren't referencing any memory for the object, this could be interpreted as a third state for boolean values accessed by a pointer or reference. (easier with a pointer as the systems are a lot more restrictive / safe for references)

But ... using nullness of a pointer as data itself sounds like really bad form. At best, interpret it as "we have no data on this".

Booleans are just booleans in any basic language (though as said, for processing speed purposes, they're coded in the smallest wordsize for the architecture, usually 1 byte). In a database, they're null only because there's an actual pointer in the database field that is null.

[Skyrunner]

I don't think you can actually make a null reference in C++...

[Helgoland]

PTW, how did this get so big so fast?

[scrdest]

PTW, how did this get so big so fast?

Derails. So many different derails.

[Reelya]

Actually, this derail into whether binary can encode gender isn't a far off derail as you would think.

Remember paraconsistent logic et al? That could be used to encode gender. How about a 4-state system of gender identification:

- male
- female
- both
- neither

Pretty much encompasses all possibilities. But, rather than now treat this as an atomic 4-value relationship, we can realize it's now two independent binaries:

- male bit
- female bit

^ this now has the advantage that you can encode people's identification with either gender separately. And it no longer has the "which sex is 1 and 0" dispute. But, it's now done away with all the funky exotic logic systems, showing that they can just boil down to binary systems by changing our assumption that the two elements are opposites.

[miauw62]

i don't actually care wether all the classes and functions in the code of my word processor are called "penis" and various immaturities, as long as the program works and doesn't contain these immaturities itself.

[Reelya]

The no-penis lobby is a different group. And yes, they very much care whether source code contains the word "penis". (Note, some commentators on the above article pointed out that some of the "sexist" examples are actually code fragments from a swear-word filter).

This thread is more about whether objects are inherently masculine thinking (object -> subject/object -> objectification), and should we abolish object-oriented programming for that reason.

[LeoLeonardoIII]

I objectify the hell out of objects.

[i2amroy]

i don't actually care wether all the classes and functions in the code of my word processor are called "penis" and various immaturities, as long as the program works and doesn't contain these immaturities itself.

Nobody cares about it... until it's your job to update it to the next version, and every object, class, and function is some variation on the words "dumb", "shit", and "penis"; then you care deeply.

[LeoLeonardoIII]

Relevant.

[Max White]

i don't actually care wether all the classes and functions in the code of my word processor are called "penis" and various immaturities, as long as the program works and doesn't contain these immaturities itself.

Still better that calling all your variables 'x', 'y', and so on, and naming every function function() or method() or doStuff()... Still, I hope if you ever write code for a living, you do it right!

[Putnam]

Writing code like that is how you keep a job.