The focus of this blog is construction-related topics. The purpose is discussion, so please feel free to comment! See Specific thoughts for thoughts from the daily life of a specifier.

17 August 2015

Tell me again, part 1

I’m sure you’ve heard the Army way of presenting information: Tell them what you’re going to tell them; tell them; tell them what you told them.

While that may be a practical way of doing some things, it has no place in construction documents. For those, we have a different rule: Say it once in the right place. I think it’s safe to say that specifiers believe this rule, though convincing those who create the drawings is difficult; the result often is that the specifications may state things but once, while it’s common for drawings to repeat things many times, and it’s also common for drawing notes to repeat what is stated in the specifications.

So what’s the big deal? Why not repeat things?
I believe the intent is good, and that everyone working on drawings or specifications simply wants to make sure the contractor knows what is needed. That’s the theory, but what really happens? In the next couple of articles, we’re going to look at unintentional redundancies.

Let’s start with specifications; it’s quite common for a specification section to say the same thing twice. Here’s an example I have used when teaching specification writing classes. It’s from a specification I found online, but the same problems are found in manufacturers’ specifications and in commercial guide specifications.
2.02    Materials

A.    Flat roof board insulation: Extruded polystyrene board to ASTM C578, Type IV, rigid, closed cell type.

1.    Thermal resistance (ASTM C518): R-5 per 1 inch of thickness.
2.    Board size: 24” x 96”.
3.    Board thickness: As indicated on the Drawings.
4.    Compressive strength (ASTM D1621): Minimum 25 psi.
5.    Water absorption: 0.7% by volume maximum.
6.    Edges: Square.   
7.    Water vapor permeance (ASTM E96): Maximum 1.1 perms.

That looks pretty good, right? Not really. Here’s the problem: Much of the information in the numbered paragraphs is already required by ASTM C578, and is, therefore, redundant.

2.02A. ASTM C578 – Standard Specification for Rigid, Cellular Polystyrene Thermal Insulation, is, as the title states, for rigid polystyrene insulation. The standard states that the insulation shall “have essentially closed cells.” The standard also states the following requirements for Type IV insulation:
  • R value: 5 per inch.
  • Compressive strength: Minimum 25 psi.
  • Water vapor permeance: Maximum 1.1 perms.

The stated water absorption is a bit of a mystery; ASTM C578 allows only 0.3 percent, while the specification allows 0.7 percent. I can’t tell if this is a typo, or if it’s measured by the same standard.

If we remove the redundancies, along with 2.02A.3 – a needless statement – we’re left with this:
A.    Flat roof insulation: ASTM C578, Type IV.
1.   Board size: 24” x 96”.
2.   Edges: Square.  
And that could be further reduced to a single statement by including the subparagraphs in the main paragrah. Instead of 9 paragraphs and 71 words, the insulation can be specified in a single paragraph using 14 words.

The usual objection I get is, “So what? What’s a few extra words? They’re correct, aren’t they?”

They are, but why are the requirement restated? Doing so adds nothing; more important, one could argue that because only those performance criteria are stated, the specifier doesn’t care about the other things required by ASTM C578, such as density, flexural strength, dimensional stability, oxygen index, the test temperature for the R value test, or acceptable defects. Part of the problem is that specifiers often state requirements that don’t matter, simply because they’re in a manufacturer’s specification.

The usual counter is, “Of course we want all that, too. The contractor has to provide it because it’s part of the standard.” If that’s the argument, then why list any of the properties required by the standard?

Another argument is that specifying those properties makes it easier to review submittals. I suppose that’s true, but again I ask, what about the other properties?

Another problem with restating parts of the reference standard is that doing so introduces another possibility for conflict. In this case, it’s quite possible that the specified water absorption is a typo. Another possible problem arises when a person unfamiliar with the standard changes the Type, say, from Type IV to Type V, and doesn’t change the compressive strength.

Virtually any reference standard contains a multitude of requirement, some stated, some incorporated by reference. Their value lies in the fact that requiring compliance with them automatically makes the entire standard part of the contract documents. Selectively restating selected parts of those standards is not only unnecessary, but it suggests that the few things cited are the only ones that are important.

Another redundancy in specification sections is created when a manufacturer’s instructions are included in the section. A simple “Comply with manufacturer’s instructions” makes those instructions part of the contract documents. It also avoids problems created by incorrect copying, and by changes in the manufacturer’s instructions.

Further complicating the issue is the fact that different manufacturers may well have different instructions. If a specification section is based on Really Great Coatings Super Stuff, which is applied at 30 mils, but you get Coatings-R-Us, which goes on at 60 mils, the specification is simply wrong. You could address the problem by specifying requirement for one product, followed by “Or other as required by manufacturer” but why not take it a step further, and simply require compliance with the manufacturer’s instructions?

Of course, your personal experience may have been that you want something other than what the manufacturer requires. If that’s the case, you may be justified in changing the manufacturer’s instructions. Be aware, though, that if something goes awry, the contractor may well blame the problem on you.

Next time, we’ll look at how specifications frequently repeat requirements stated in Division 00 and in Division 01.

© 2015, Sheldon Wolfe, RA, FCSI, CCS, CCCA, CSC

Agree? Disagree? Leave your comments at


  1. Thank you for discussing concerns about how to use industry standards in construction specifications. One of the reasons for errors is that few specifiers and contractors have copies of the thousands of standards used in construction. I would welcome your thoughts how small architectural, engineering, and construction firms can get affordable access to standards.

    1. That's something even large firms have trouble with. I work in an office of about 140, and we cannot afford to buy every standard. However, ASTM's "ASTM Standards in Building Codes" is affordable, at $1,000 in hardcopy to about $1,500 for the online version. As the title suggests, this reference library contains the vast majority of ASTM standards used in construction.

      Many standards are free downloads, and many project representatives purchase standards and give them to the designers they work with.

      While it is extremely difficult for a designer or contractor to have copies of everything, there is no excuse for manufacturers or subcontractors to not have copies of the much smaller number of standards that apply to their work.

      Why should anyone have to pay for building codes, or standards they are required to follow? They are, in essence, laws, so they should be free. Look up Carl Malamud and his; he has been fighting for years to make state laws available at no cost - and winning.

    2. I am a small (two person) spec consulting firm and I maintain hundreds of up-to-date standards that I need for my work. The expense is simply the cost of doing business. Any design, construction, or manufacturing firm, large or small, would be neglect to not have current standards applicable to the conduct or operation of their business. It would be like doing their taxes based on outdated tax laws?

  2. 1. Totally agreeable and sensible. 2. An office may be able to afford to buy the books, but they cannot typically afford to read and understand them. Most standards reflect a political process that in turn reflects the interest of both the "common good" and the interest of the stakeholders that sat on the committee that developed the standards.. These are subtle details that seem unimportant (and often unintelligible without a lot of background) in the absence of a lawsuit or other serious conflict. Your example mentioned measurement methods, typically something understood by few of those who write or are governed by the spec.

    1. Excellent points, Leo. I've been a specifier for nearly thirty years, and I still haven't read - let alone understood - but a relative few of the standards I use. Soon after I became a specifier I realized I would never know everything I need to know to do my job (, and each day the gap widens.

      And, as noted in "Faith-based specifications", there is little rational basis for criteria established by standards, making them even harder to understand.

  3. I welcome the effort of Also, I applaud NFPA for making their standards available without cost.

  4. This is an issue I am working through right now when I review designer contract documents for work on our Memphis UTHSC campus. Another related problem is the specifier will list reference standards applicable to the section in Part 1, many of which are NOT related to the section. I would prefer the designer not list anything if they are not willing to do the research to ascertain if the reference does indeed apply to the specification section. Another issue in Part 1 is "Related Sections", also suffering from the same malady. I am all for coordinated documentation but when the specifier fails to confirm if those sections listed are actually related then why list them at all?

    1. Why, indeed! Although the References article is often used, it is not universal. With SpecLink and other automated specification software it's no longer a problem to have a correct list of the references used in section, but many specifiers do not include that article unless required by the owner.

      The Related Sections article has a somewhat different problem in that judgment is required to compile that list. Too many people don't understand that this article should be limited to those things "the reader might expect to find in this section", and instead generate what sometimes appears to be a table of contents. Taken literally, "related sections" would include all spec sections and drawings, but that would hardly be useful.

      Thanks for the comments, Hans!

  5. suggests that the cost for the Construction-related ASTM standards is almost $3,000. That's a huge pill to swallow for a small architectural practice, on top of whatever one might pay for a subscription to a "prepared" master specification product.

    I agree that errors in citing a reference are a problem, but to use the whole ASTM nomenclature is a challenge. Unless you have the book with you, or have memorized the ins and outs of what you specify, how does one know if the item in front of you on the job site is the Type III or the Type VI version? Perhaps it's printed on the container or package ... and we hope they got it right. For me, a little bit of plain English, As Long As It Is Correct, goes a long ways.

    I agree that "References" is often overused. Many specification preparers seem to think that the more standards mentioned, the more likely someone will be able to "catch" the contractor ... bad approach.