Positive or Negative

Question

Positive or Negative

Viewing 15 posts - 16 through 30 (of 129 total)

You must be logged in to reply to this topic. Login to reply

RonKyle SSC-Dedicated Points: 31556 More actions · Answer 1

I'm not sure I'm consistent. For items that are look ups of one kind or another (that would be considered dimensions in an OLAP environment), I use an Active flag with a default of 1 (Yes). But for main items that would be akin to fact tables in an OLAP environment, I use Deleted with a default of 0. But somehow using Inactive and NotDeleted doesn't seem to work.

tom.w.brannon Hall of Fame Points: 3976 More actions · Answer 2

The convention for the group I came into when learning SQL Server is *_is_active_indicator which makes for some very long names for some very short columns

call.copse SSCoach Points: 18048 More actions · Answer 3

I would tend to use Enabled or Deleted in most cases, probably mostly on the flag exceptions basis as described so far. I don't mind having a default of 1 for such things when it seems to make sense though.

Maybe Active would work for something like a list of servers that could be active or decommissioned. Inactive would strike me as trickier to mentally parse in queries, maybe that's just me.

brett@zerozero.net SSC Veteran Points: 245 More actions · Answer 4

There is some study that I read once that indicated that as English speaking humans, we think more in the positive sense, So I always try to use Positive terms for everything.

I do think it makes complex logic much easier to read and understand having had to decode both in my career..

This is more important in languages like c# where you have true Boolean logic but ultimately our databases are used by these guys.. so make their lives simple

if (active && !deleted)

{ //do something

}

more natural to read and understand as English speaking humans than:

if (!inactive && !deleted)

{// do something

}

david.dilworth Right there with Babe Points: 780 More actions · Answer 5

As a C# developer I believe it's better to show the intent of the code or schema with a positive naming convention. Therefore I'd definitely favour the "IsActive" style. I think this leads to clearer logic when faced with the double negative implied by "NOT InActive". Here are some examples from our code: "IsOn", "IsComplete", "IsCurrent", "IsAvailable", "IsDeleted".

Megistal SSCrazy Eights Points: 8787 More actions · Answer 6

I go first with the way the business model works and run.

If a ties occurs, I go with the most easiest way the developer will uses it.

If a ties still occurs, I use the positive form to avoid double negative way, choosing simplicity.

This summarize

brett@zerozero.net SSC Veteran Points: 245 More actions · Answer 7

I understood that the question was about philosophy of using positive or negative for the attribute names

This naming convention if should it be X, IsX, DoILikeX or X_Flag or whatever is a totally independent matter (in my view).

angel-626714 Valued Member Points: 56 More actions · Answer 8

I'm for the inactive. I would also have a column for the date and time it became inactive.

Phil Parkin SSC Guru Points: 246956 More actions · Answer 9

david.dilworth (10/23/2015)
As a C# developer I believe it's better to show the intent of the code or schema with a positive naming convention. Therefore I'd definitely favour the "IsActive" style. I think this leads to clearer logic when faced with the double negative implied by "NOT InActive". Here are some examples from our code: "IsOn", "IsComplete", "IsCurrent", "IsAvailable", "IsDeleted".

+1

The absence of evidence is not evidence of absence
- Martin Rees
The absence of consumable DDL, sample data and desired results is, however, evidence of the absence of my response
- Phil Parkin

Xavon SSCrazy Points: 2280 More actions · Answer 10

I use the positive (which is problematic because a previous developer used the negative).

I prefer 'Enabled' to 'Active', but since 'Enabled' is a keyword and I consider using keywords as column names a sin, I go with 'Active'.

I am also not a fan of 'IsActive', because I find the 'is' part redundant, and because it is too easy on a cursory glance to read it as 'InActive'.

RWillsie SSC Enthusiast Points: 144 More actions · Answer 11

U.S.A.F. Instructors pounded it into my head many years ago to never use "negative logic". To this day I find it offensive when having to deal with a language or app that sees "-1" as "True".

Positive please...

Matthew Pride Hawkins SSC Rookie Points: 28 More actions · Answer 12

We use negatives, IsInactive and it does make it harder to explain to folks but unfortunately you do get used to it.

Bill Talada SSChampion Points: 11954 More actions · Answer 13

Some developers ain't never did nuthin' with nots and ain't not never gonna will.

My complaint is using the math terms of positive and negative when we are talking about a state being asserted or negated.

And secondly I would warn about the complexity of using several columns to define a state of a row.

Andy Painter SSCommitted Points: 1820 More actions · Answer 14

I'd go with the positive approach: Enabled, Active, Deleted etc.

Much easier to understand

david.wright-948385 SSCarpal Tunnel Points: 4041 More actions · Answer 15

angel-626714 (10/23/2015)
I would also have a column for the date and time it became inactive.

Denormalisation alert 🙂