Meeting minutes
Decision tree
Julie: Monday at TPAC, we will be reviewing where we're at with AG and Tuesday will be looking at conformance models
Julie: The draft of our sprint is coming along, but we need more time - we will be updating AG on this today
Julie: Yesterday, I added a draft decision tree as a start for today's conversation.
<julierawe> https://
Julie: I just want to get gut reactions to the decision tree see if this is on track. It's based on our recent conversations.
Julie: Does this feel like the first piece should be a requirement, or is it a method?
JohnR: I think it's a requirement
Laura: I think it's a requirement
Julie: To everyone else - having it start with a requirement, is that okay? The question makes sense, but is it a method of how to do something?
JohnR: An important thing to understand is that the link from requirement helps people understand whether or not they have implied meaning, which is essential.
JohnK: It doesn't seem to fit my standard way of doing outcomes because an outcome would be that the meaning is made understandable by (fill in the blank) - the outcome is the outcome, so the language seems a little wrong - the outcome feels like a method from my perspective.
JohnK: The outcome is that the implied meaning is surfaced.
Julie: We want the end result to be understandable to everyone, so we present two paths, which is what questions 2 and 3 get at - which is how to handle implied meaning.
Julie: The decision tree leads to the outcome, but what we have to decide is if the first step is a requirement or a method.
JohnK: To me, there are 2 steps - one is determining whether you have implied meaning, and the next step is what you do with it.
Julie: Yes, it feels like 2 steps - determining whether you have it and then what to do about it.
Julie: Step 1: Does it contain implied meaning. Step 2: Is it essential to use this language? This is about the author's preference. Step 3: If the answer is yes, then are you making an explanation available?
JohnR: For number 3, I am not sure what you mean by "uses" related to second method. Why is the yes for number 3 different from steps 1 and 2?
<kirkwood> an example i think of implied meaning “policeman,” “cop,” and “The Man” is this not literal language?
Julie: The first one is do you have implied meaning and the second one is do you want to keep it? Step 3 is do you have an explanation of implied meaning? We could just have a one-step decision tree and recommend that authors avoid non-literal language and start with step 3, where people have to explain implied meaning.
Frankie: I was thinking about focus visible - this kind of follows that same pattern. First, is it there, and then you go from there. I don't understand the comment of taking us back to just step 3. We have to determine whether it's there and then what to do about it.
Julie: We would not know if someone decided to take out nonliteral language and replace it. The second step is for the author to think through. The only thing that is testable is step 3.
JohnK: Implied meaning in literal language - I gave an example in IRC - policeman, vs. "cop" vs. "the man".
Julie: To me, "cop" literally means "police." It might be a less formal name, but I am not sure the example is an implied meaning - maybe something like, "the man," "the fuzz," etc. You couldn't necessarily go to the dictionary and look up the word.
JohnK: In the dictionary "cop" might say "informal."
Julie: Yes, but "informal" is not the same as "implied."
Julie: For implied meaning, it would be things associated with a word, but not the definition of the word - situations where we might lose some readers who would not understand what we are referring to.
<kirkwood> was just trying to put a real world example in front of us
Julie: If we look at the decision tree, the first step is recognizing if we have implied meaning, the second step is deciding whether to keep it, and the third example is how to deal with the answer to question 2.
Makoto: I have no objection with the decision tree. Do we have to provide explanation every time we use words with implied meaning. One example in English is "piece of cake" - would we have to provide an explanation, even if we know the intended users would understand it?
<kirkwood> Maokota gave “piece of cake” example and asked if need to explain that
Julie: Last week, we talked about intended audience. For example, a car website, users might understand what the parts of an engine are, but if you said, this engine "purrs like an engine", people who are literal might not understand that.
<kirkwood> those are idioms
<kirkwood> those are common idioms
Julie: If we were to carve out an exception, for example, I expect everyone in my audience to understand "idioms," then I don't have to explain it. That would mean that the author would never have to explain anything. Julie - I think there's a difference between jargon and idioms. For example, jargon would be that accountants understand accounting
terms - that would be jargon.
Julie: Does anyone have a reaction about the difference between jargon and and idiom?
Lauren: I like Julie's definition and we should document it because the question will come up?
Len: I like the idea of documenting idioms, metaphors, analogies, etc. so that people are on the same page.
<kirkwood> in my past gov’t work was a standard (usu in style guide_: no idioms or analogies
Lauren: Maybe we could put this in the user need.
Jan: *correction* - where I have "Lauren" above - it should be "Laura"
<kirkwood> some people think of style guides just as colors and fonts
John: the style guide should include no analogies, etc. You can't just point to a style guide and get credit. The style guide has to address cognitive issues
Julie: I don't necessarily want to describe what a style guide "must" say, but we could point out things that are common in style guides.
JohnK: the common things I have seen in style guides are more visual issues - like the type of fonts, etc. They don't necessarily address how to effectively create content that is usable by people with cognitive disabilities.
<kirkwood> will it change the meaning? is how i’ve seen it
<kirkwood> +1
<kirkwood> +1 to Laura
Laura: Maybe just define "essential"
<Laura_Carlson> essential: if removed, would fundamentally change the information or functionality of the content, and information and functionality cannot be achieved in another way that would conform
<Laura_Carlson> https://
Julie: I am thinking about the branding example - where your brand might be using puns or clever language - non-literal language would not necessarily change the meaning, but it might change the brand.
<kirkwood> the tone may be a part of the ‘brand’ and part of the experience thus essential
Julie: There are instances where people might want to use non-literal language.
Frankie: One of the things we talked about in the Internationalization group - we have some feedback from the community group about exceptions for art / literature, etc.
Julie: If you have an exception for "literature or art," we would still want you to explain non-literal language, correct? Otherwise, we couldn't say that you've met this outcome.
Frankie: If authors post an online poetry journal, they are not going to explain every line. However, if it's an educational site and we are teaching students how to understand poetry, then we might explain a Robert Frost poem, for example.
<kirkwood> “purpose” was a key concept Frankie brought up
Julie: Using the poetry example, it would be "essential" but if you decide you don't want to explain it, then you would not meet this outcome. Would we then say that there are instances where it would be okay not to explain implied meaning?
JohnK: Frankie brought up a key word - purpose - maybe this is what we need.
Frankie: What about testing sites where you would want understand what students know.
Julie: Would advertising be an exception? Are there other exceptions to consider?
<LenB> +1
<Makoto> +1
<Frankie> +1
Julie: Please put in a +1 if the current draft of the decision tree is okay, -1 is it needs more work and 0 means I am not sure:
<Jan> +1
<kirkwood> 0
<JohnRochford> +1
<Laura_Carlson> +1
<Kimberly> +1
Julie - before next week, let's think about the exceptions idea - what would qualify for an exception and what would the rational be for them?
<kirkwood> I would like examples, very helpful
Julie: We have talked about creating a wiki where we have examples in many languages - maybe we could have a quick English example and then prominently say that we are creating a wiki, or would putting English examples be problematic because we're giving too much preference to English?
Frankie: Right now, I have no preference of where we put examples, but I am trying to think about where to put in some work this week. Would it be helpful to come up with examples?
Julie: We need some help in the test section, if you have time, please think about either building out the 2 methods or some tests.
JohnK: I find that examples are good in supporting understanding so that we can even develop tests. We can't develop tests unless we truly understand what we're testing.
<Laura_Carlson> Examples to help us think about methods and tests: https://
Julie: Yes, but to keep this from getting too long, we just need to decide how to handle the international scope, so should all of the examples live in the wiki, or can the English examples be in the document, but other languages be in the wiki?
<Laura_Carlson> /www.tpgi.com/maturing-your-accessibility-program-kpis-and-strategies-for-success//docs.google.com/document/d/1P7fOyEPVlqf1aXuJY0SO9LeC-E7EZllg/edit#heading=h.4cabv7fg6orv