Numerical comparison of 14.5 between 2009 vs 2018 1

[3DDave]

As an estimate:

Word Count:
2009 - ~40,000
2018 - ~60,000

Sentence Count:
2009 - ~2000
2018 - ~3000

The sentence count is rather significant as each sentence is essentially a rule. A 2018 user needs to memorize nearly 3000 rules, 50% more than 2009. My eyeball estimate is this additional rule count is about the same as the total number of rules in the 1982 version, per my recollection. To borrow Mark Twain's estimate method, there would have been no rules at all in 1991.

Some of this growth is from adoption/duplication of rules for Model Based (CAD), which I suppose is nice, but is mostly going to be automatically covered by the CAD software developers and should have been solely in Y14.41 - Digital Product Definition Data Practices. There's no good reason for them in Y14.5.

What I would like to know about the rest of the increase is if the utility of the standard has gone up in proportion - that the vast increase in rules has made it far easier to understand or to cover a proportionate number of cases that were previously out of reach by any other method.

Comparison: '2009, 3.3.3 Datum Target Symbols had, basically, 5 rules. As far as I know this was the same back to the 1982 version. The '2018 6.3.3.1 (note the additional level) Datum Target Symbols has 8 rules, 1 of them specifically for Digital Product Definitions.

In the future, is the desired goal to get to 5,000 or, eventually, 10,000 rules, or is it just an acceptable part of adding more special cases? Going that way makes the picture-book format a better choice.

Is there consideration for even more duplication in this standard for topics that are part of other standards?

Is there any plan for this expansion and does the committee have the tools to provide a clear mapping from one version to the next rather than the woefully inadequate PRINCIPAL CHANGES AND IMPROVEMENTS?

Most annoying is the nearly 100% rewrite that prevents simply flagging the changes the way most revised documents can have change-bars.

I feel sorry for those who are supposed to get through this mass to get their job done. They certainly cannot memorize 3000+ rules and are almost certainly just going through the pictures to find a situation similar to the one they think they have.

 

[Nescius]

3DDave,

I generally agree with the sentiment of your post. In my opinion, the folks in charge of the standard deserve hard criticism; the standard is simply not very good. I will temper my rebuke a bit, though; it is not in the nature of a committee to produce art. Note, 3DDave, I'm only riffing on the issues you've raised; I'm not trying to put words in your mouth. Furthermore, I haven't even read most of the newest standard. Why? One of your closing statements sums it up:

I feel sorry for those who are supposed to get through this mass to get their job done.

I'm practically a geometric tolerancing evangelist and even I feel like the standard is bloated and, even worse, unstable and inconsistent. A shocking proportion of 14.5 application out in the wild is incorrect. Most engineers and machinists know next to nothing about it. The standard being absurdly expensive is just icing on the cake. 14.5 has been around for decades and its roots are still very shallow outside of sheltered enclaves. The committee is only making that situation worse.

Some might laugh at this, but I'd say the situation is so dire that a large industrial player could make 14.5 obsolete by building an in-house standard, gently bullying their suppliers, and making it open source. If, big IF, the thing they built was truly artful, it could push 14.5 out.

 

[Burunduk]

How does all that compare to ISO GPS?

 

[3DDave]

Bad solutions don't get better because there are worse ones. Tolerating bad solutions pave the way for worse ones.

 

[Burunduk]

Are you sure that each sentence is a rule?

 

[powerhound]

When stubborn engineers say that they don't know what the word "Shall" means and imply that it could be an option, things get ridiculous. Most of us know that when a note says that a part "Shall be anodized" or "shall be powder coated" no one thinks twice about it. When the standard states that datum features "shall be controlled" suddenly it's a mystery. Now, starting sometime around 2013, we have a section in every standard that defines the word "shall" as establishing a requirement, just like what it's always meant. Also there's a definition for should, will, and may. None of those definitions are any different than when they are used in everyday language. This is the kind of ridiculousness that really gets under my skin. Count the words in that section of the standard that defines those words and know they are there because of folks reluctance to just get with the program and stop making the committee do these little things.

John Acosta, GDTP Senior Level

 

[3DDave]

Curiously, one shift is from the word "must" in 2009 to "shall" in 2018.

At the same time, in legal circles/government, there is/was a move to stop using "shall" and instead to use "must".

https://www.plainlanguage.gov/guidelines/conversational/shall-and-must/

"Second — and related to the first — it breeds litigation. There are 76 pages in “Words and Phrases” (a legal reference) that summarize hundreds of cases interpreting “shall.”

So it's a problem not limited to Y14.5

 

[axym]

Hi All,

I would have a very different take on words and rules in the standard. I agree that the word count is too high and that the value of the additional words can be questioned. But wouldn't agree that each sentence is essentially a rule, and that there are too many rules to memorize. There are a lot of words and examples, but there aren't enough rules.

Part of the fun of participating in Y14.5.1 is that the approach to words and rules is much different than in Y14.5, because it needs to be. The challenge is to distill the sentences in Y14.5 down into actual rules that can then be mathematized.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[3DDave]

For my estimated count - the question is can this sentence be eliminated from the standard and that no one will argue it is required?

If it isn't required, who argued for it to be put in? Not looking for names, just that someone did argue for it.

If no one argued for it to be there how did it get included?

If it was argued for as necessary and voted that is had to be included, then it is a rule intended to govern or regulate something and is intended to be used to win some argument.

For example, someone felt that this was required: "The spotface diameter is the distance across the flat and does not include any fillet that may be specified" possibly due to having had an argument about just how to reject a spotface and they wanted to more easily win that argument in the future.

I'm sure someone is itching to deal with spotfaces that aren't circular to handle the light cleanup of surfaces surrounding slots in rough castings.