Airtable Accessibility

I am about to embark on a large research project with the Royal National Institute of the Blind and one of the requirements is that the format we hand over our research should be fully accessible. Given there is going to be a lot of research findings I was hoping to use Airtable, but is it accessible?

Thank you
J

Funny you should ask - the University of Michigan’s Digital Library Accessibility Team dropped into this Community in September making broad and sweeping accusations that Airtable was “Inaccessible”.

I challenged their test methods and conclusions over a three-week timeframe all while testing their own library content against the WCAG standards.

Their own sites failed, of course, and while it was not in debate, Airtable had some issues. UofM must have been a bit embarrassed because it appears the entire thread - which included test results and a variety of comparisons to various WCAG standards - has been deleted from this community.

I’m interested in this question too, albeit in a less immediate way. What exactly does “accessibility” mean in this context? I found this article on accessibility at the W3C:

https://www.w3.org/standards/webdesign/accessibility

Are the recommendations there the sort of thing that we’re talking about? If so, I’m wondering how many database applications meet those standards.

William

Yes, that’s it. What “accessible” in this context means is that a person with a disability (typically, but not always, a user of “assistive technology”, such as a screen reader) has the opportunity to benefit equally from the web site, application, document, etc. In order for this to be true the website/application developers have to employ techniques to make it work with assistive technology and without a mouse. Color contrast needs to meet minimum standards. Audio transcripts must be provided. Video captioning and audio description must be provided. The current standard that is most widely used is WCAG 2.1 Level AA

Notwithstanding Bill.French’s comment above, whether or not a university’s own website is fully accessible has no bearing on their procurement processes that may require outside providers’ products to be accessible. When a website or product is not fully accessible, an organization SHOULD have an “Equally Effective Alternate Access Plan” developed to ensure that those with disabilities can enjoy equal benefit. Testing someone else’s website doesn’t tell you whether they have an alternate access plan but that plan SHOULD be mentioned on the affected website.

Airtable SHOULD have already tested their products and SHOULD report their accessibility status in a “conformance report”, informally known as a VPAT (Voluntary Product Accessibility Template). Some companies publish their VPATs publicly and some do not.

I was looking at Airtable as a way to track and display the accessibility status of all the products we use at our institution but lacking a VPAT I have to look elsewhere, even though we do have legacy inaccessible products in use here too.

Michael Cyr

Indeed, but that was not the context.

UofM’s accessibility “library” was not inquiring or commenting on the accessibility of Airtable for procurement purposes. Instead, they were [apparently] trying to justify their existence (and perhaps their funding) by finding tools that may not ideally meet all accessibility standards and then publishing broad and sweeping [headlining] statements of failure to comply.

I have no problem with calling attention to Airtable’s lack of compliance, but doing so with sensational headlines in a customer-facing forum and all without ever attempting to contact Airtable directly, calls into question their motives and also opens their own content to direct scrutiny.

Given that posture, full accessibility and compliance for their own “digital library” are very relevant. One must challenge and hold accountable the very organizations that claim high ground and cast long-lasting negative dispersions about commercial products, and especially when their own backyards are littered with inaccessible content whose topics ironically make claims that all content must meet accessibility standards.

With specific regard to WCAG 2.1 and VPAT, I cannot comment (not an Airtable employee). However, compliance requirements have predicates and I assert that there are some aspects of a database development tool that qualify for a few of the well-documented exemptions. And I’m sure they also have some work to do as do almost every SaaS vendor on the planet likely has room for improvement.

Without the original posts it’s hard to judge. I work with higher education accessibility professionals all over the world and I have yet to meet anyone who goes out of their way to identify and pounce on companies whose products don’t meet ALL accessibility standards. It’s a truism in our world that there are no 100% accessible digital products. What we do have to do, according to the Federal Dept. of Education, is confirm vendor’s claims of accessibility. It’s ludicrous that the burden isn’t on vendors to publish and stand behind their claims. Thousands of institutions have to duplicate their analyses of the same products! Hence there have been attempts to put together libraries of our findings. Again, without the original post I have to withhold judgement.

I did do some basic testing of airtable and it’s much worse than a lot of other products. However, anything with a tabluar display takes extra work to make accessible. 10 years ago, very few SaaS products passed even basic accessibility testing. The landscape is much improved but there’s still a broad spectrum of compliance. Airtable doesn’t seem to have any public statement or commitment to accessibility. I have requested their VPAT. If I get one I’ll update this thread.

Mike

Precisely. Why would they retract the post and the history? The only plausible explanation is they likely realized they were overly harsh and their approach was flawed.

Please be specific - as compared to what exactly? Are your tests of a specific Airtable solution or are they of the tools and processes provided in Airtable to make solutions?

That’s not the only plausible explanation. Many who test for accessibility withhold their results from the public due to fear of litigation. Whether or not one is “correct” has no bearing on the cost of fighting such suits.

I tested the basic airtable DB table view/edit screen. I compared it to the WCAG2.1 level AA standard, section 2.1, Keyboard Accessibility, using the NVDA screen reader. It’s completely inaccessible, that is,keyboard-only control does not provide corresponding screen-reader accessible text that matches what is visually displayed. I test a lot of applications. Total lack of keyboard and screen reader support is unusual these days. I also tested the read-only view of the table from a shareable link. Same lack of support.

I’ve also received a response from airtable. They do not have a conformance report. I’m asking for clarification about what they DO know about the accessibility of their DB product.

Yep - I get it, there are many good reasons to follow these standards and attempt to achieve compliance. As I said, Airtable probably has a lot of work to do.

Are you saying that not a single person with accessibility challenges of any kind whatsoever could never use any aspect of Airtable?

That’s a scary statement to any current or future buyers considering Airtable. And yet, WCAG 2.1 AA is darned near the most stringent test you could possibly administer, right?

The publication of WCAG 2.1 does not deprecate or supersede WCAG 2.0. While WCAG 2.0 remains a W3C Recommendation, the W3C advises the use of WCAG 2.1 to maximize future applicability of accessibility efforts.

Furthermore, 2.1 is the standard you would advise if you want to dovetail accessibility recommendations into the [distant] future, right?

I’m sure it’s a good thing to plan to meet 2.1, but I just want to understand - today you are measuring Airtable’s accessibility with a “standard” that really hasn’t been adopted yet, nor is it likely to be adopted in the near term, right?

WCAG 2.1 is not required by the revised US Federal Section 508 standards. The US Access Board would need to start an official process to update the revised standards to reference the updated WCAG.

You said earlier that other tools like Airtable are totally accessible. Can you name them and provide the test results so I can advise my clients?

Lastly - and to help us all understand this better - tell us about the distinctions (both legal and standards-based) between these six “web” environments. Does WCAG v2.1 apply to APIs? Web Apps? Database tools? SQL language? What are the exceptions embodied in the WCAG overall?

  • Web Site
  • Web App
  • Mobile App
  • Database Management Tool
  • SQL
  • Database API

I should have said that it’s complete inaccessible to people who are blind or visually impaired. BVI is the most common source of accessibility complaints but obviously there are others. I could test contrast ratios for example… but not bothering because failing the most basic keyboard test means that we can’t adopt the tool.

No, level AAA is the most stringent of the WCAG standards. The US section 508 (which applies only to federal agencies) is based on level AA. The level of stringency is really irrelevant as the courts have pretty much agreed that level AA is reasonable and they continue to hold institutions and organizations accountable to it under the ADA even though the ADA has no regulatory framework in the area of digital accessibility.

If a manufacturer could show me that they meet 2.0 AA and not 2.1 AA then I’m not going to be too concerned. Remember I said that we accept that nothing is 100% accessible. There’s other precedent in the ADA that encourages a case by case accommodation strategy. However, if you provide a public accommodation, it’s in your interest to reduce your risk to lawsuits, no? The courts have already said that the evolving nature of these standards is not an excuse to ignore them. SCOTUS just refused to hear Dominos Pizza’s appeal based on such a claim.

I did not claim that there are accessible alternatives to airtable. There are however many web-based applications (SaaS) that are accessible. Google Docs for example, are pretty good and getting better all the time. Microsoft is doing a good job with Office 365. I don’t have an accessible alternative to airtable but that doesn’t mean there isn’t one. It is the responsibility of everyone who creates a public accommodation, under the ADA, to ensure that the technology is accessible… same as if you build a store, it needs an access ramp.

There aren’t exceptions in WCAG because WCAG itself is not a law. There are two exceptions built in to the ADA. “Fundamental Alteration” and “Undue Hardship”. Financial hardship rarely qualifies as undue hardship because the Feds look at the cost of accessibility against the entire organization’s expenditures. An undue hardship for example might be that there really is NO accessible alternative to airtable. That would be your burden to prove though and it’s a pretty high bar. However, even if you make the undue burden case successfully, you still have to provide an “Equally Effective Alternative”. I have a client that publishes 600 page inaccessible PDFs of their Board Materials. The PDFs are produced by an app. They spent a month trying to find an accessible board book app and couldn’t. So they publish the inaccessible PDFs, along with a note that they will make the source files available to anyone (qualified) who needs them.

“Fundamental alteration” is pretty straightforward. For example, you don’t need to make a driver’s eye exam accessible to a blind person because doing so fundamentally alters the purpose of the test, i.e., to determine whether someone can see well enough to drive safely. Similarly, flight schools don’t have to make flight simulators for student pilot use accessible. because someone with a disability that would affect their use of the simulator would not be qualified to earn a pilot license.

The distinction to keep in mind is whether you have created a public accommodation or not. The Dominos Pizza website is a public accommodation. The development environment that their web developers use is not a public accommodation. Hence a member of the public, who doesn’t have access to the Dominos development environment has no standing to make a claim of discrimination. However, an employee who needs to use the development environment could well make a claim that they suffered discrimination if that environment creates barriers to their job performance, advancement, etc.

Here’s an interesting question though… what about someone with a cognitive disability that prevents them from successfully developing DB queries using SQL? That would probably qualify as a fundamental alternation exception because the nature of the work is fundamentally complex and that’s reflected in SQL itself. It’s not the technology that’s creating the barrier, it is the job itself. The ADA requires Equal Opportunity, not Equal Outcomes.

The above is is no way legal advice. Consult your counsel if you have questions.

Michael Cyr

@Michael_Cyr,

This is fascinating. I love these deeper explanations and welcome additional sharing of these stories to help Airtable developers do their best to create better accessibility.

Ah, so the vast apps that are not publically accessible – pretty much every Airtable app – are fundamentally exempt from accessibility standards unless such private accommodation is central to the work of employees who may require accessibility features to sustain parity with workers who do not require accessibility features.

Very interesting.

One last question - accessibility test results are generally gated to some degree by OS, browser version, and accessibility plugins. Just curious - did your tests exclude all the advantages available from these accessibility features?

Oops! I lied - one more question.

This makes sense and I can see how something like a large office suite (like Office365) would need to make this accommodation. But to be clear, when you tested Airtable were you testing the use of the platform to build a database or use a database?

Typically, there is no “public accommodation” extended to anyone concerning the crafting or configuration of a solution. However, once a solution is placed into production, many features are limited and sharing is yet again, vastly private. As such, the features of a solution can vastly vary as compared to the development platform itself. Does this matter in applying WCAG standards?

Yes, in my opinion, if one is using airtable to build an internal app only used by employees, then you have less exposure - until an employee shows up who needs the app to work with assistive technology. The problem at that point, if you’ve built the app on a fundamentally inaccessible platform, is what to do? Are you going to instantly move the app and its data to another platform? You can assign the employee other duties but risk a lawsuit. If a person with a disability APPLIES for a job, gets an inkling that airtable use is involved (either as developer or app user) and is turned down, for any reason, they may well make the claim that it was their inability to use airtable that cost them the job and sue. Hence, for a small company with a few employees the risk is much smaller than for a large corporation or institution.

We have 8000 employees and 50,000 students. We consider anything that is all-student-facing as constituting a public accommodation. However, if a technology is used by just one course, and there are no students with a disability in the course, we may give it a temporary exception. However, because students can join a course during add/drop, the faculty is in big trouble if such a student joins and the technology is not accessible. In that case there are two options 1) immediately provide an equally effective alternative (has to be approved by our Equal Opportunity office or 2) remove the technology from the course for ALL students. Given that courses are planned and developed months ahead of time, it’s really not practical to wait until a need arises.

Yes, I use the latest versions of Windows, MacOS, Chrome, firefox, Voiceover and NVDA for testing. I am not aware of any browser plugin or OS feature that fills in missing metadata and/or structure to makes a non-accessible app work with assistive technology such as keyboard-only navigation or a screen-reader. I believe that there is a product that the app owner might use as a sort of http proxy to sit in front of a non-compliant app and rewrite its output on the fly. However, it requires separate configuration for every page type/template of the underlying app and I don’t know if it can fix page structural issues or only add missing metadata for objects.

Good question… use-case context definitely matters. I tested database table creation and data entry, and consumption of a publicly shared db table view. Whenever testing against WCAG, or any standard, test results should include the functionality that was tested, if it was less than everything. We often see, especially when a company is in the process of fixing their product’s accessibility issues, that the public or end-user facing functionality is fixed first, while app configuration or administration is addressed later.

Does airtable include an API that would allow a developer to write their own user interface from scratch? If so, then the developer is in control of the interface and could theoretically make it accessible.

Excellent assessment! It’s nice to know the bounds of wiggle-room but also the risks that can lead to a lot of pain.

Yes, and companies such as Stacker have done exactly this.

I did hear back from airtable and they do not have any accessibility conformance information about their products, nor a timeline or roadmap, to share. In my experience (which is very broad) this puts them way behind the curve and while they said they are interested in accessibility, their actions speak otherwise.

Michael Cyr

1 Like

There are many aspects of Airtable that fall into this pattern, and while there is little point in defending their [seeming] lack of resources or commitment for accessibility, I can certainly defend any business who has created an awesome tool and then got blindsided by acceptance and demand for their product. It’s not easy to transition from cool tool to perfection at scale, and especially where organizational adoption is gated by legal risks.

And what “curve” would that be? The curve of accessibility support for mature products and services?

I must ask, historically speaking, what was the ramp of WCAG adoption for products similar to Airtable? Was it as swift as your comments intimate it should be?

How many decades and billions in profit have been earned by Microsoft for its flagship Excel product for example - and yet, they still struggle with meeting WCAG standards. Anyone can build an Excel spreadsheet that does not meet standards and requires more work by the sheet creator.

For every accessibility standard that Airtable does not presently support, I can show you modern, well-respected tools that also do not meet WCAG v2.1. The accessibility debate is an interesting one, but meeting all standards seems to be fleeting at best and challenging even for mature products with virtually unlimited resources.

Airtable - like many software companies - probably view this challenge from purely economic perspectives.

  • Do we direct significant resources to meet WCAG standards?
  • Or, do we continue to address key market demands that will sustain our competitiveness and continue to satisfy our core user base?

At some point, they need to do both. But for now, they are apparently telegraphing a strategy that is not unlike most solutions that are experiencing early success and adoption by every-day business and organizational experts who need code-free tools to do their best work.

I suspect they also gauge their strategic decisions concerning accessibility as it relates to the target user demographic. By-and-large, demand for their product is vast among personal and small business users and I have a feeling that this customer segment is where they derive their core revenues. Sure - they want pervasive adoption in the enterprise and the big [potential] revenues that come from that segment. However, a tool like Airtable is not ideal for large businesses and accessibility is WAY down the list of reasons why it is not presently suitable for enterprise-wide adoption.

Any use of Airtable in enterprise settings is likely occurring for the same reasons dBASE was dragged from the hobbyist’s home PC into the workplace in 1985. It is no different than iPhone (one) making an appearance in corporate sales teams in 2007 where Blackberry was well-entrenched. It is no different than the procurement by enterprises to wrestle web technologies from the arms of global small businesses in the late 90’s and which fueled the dot-bomb era.

Neither dBASE nor iPhone nor HTML were suitable competitors to dominant technologies in use at enterprises at the time they made swift penetration. Airtable (and other code-free tools) are no different - they are disrupting the status-quo in enterprises; crashing the establishment-era application party and the defence mechanisms are beginning to kick in.

It’s unfortunate that an organization like yours is hamstrung by litigation risk from adopting such a great product like Airtable.

Agreed… it sucks that it’s not airtable that’s on the hook and instead institutions and companies that chose to use or base their products on airtable, or other inaccessible tech. Hence why all entities should have “accessibility review” as part of their technology adoption/procurement processes.

Yes, that’s the curve.

When accessibility was new (15 years ago) almost no one took it into account when building their products. Generally those extant companies/products started taking this all seriously 5 to 10 years ago. However, it’s still not unheard of for startups to ignore accessibility while trying to prove their business model. Problem is that more and more large buyers ARE considering accessibility conformance. Hence, accessibility is becoming a business case. I’m part of a movement building accessibility review processes and tools for higher ed, where lawsuits are rampant (see the recent MIT/Harvard/California Community Colleges/California State University/etc. cases and settlements).

There will likely NEVER be a product that 100% PREVENTS a content creator from doing something inaccessible. That should not be on the product manufacturer if they have given the content creator the tools to make their output accessible. For example when Word/Excel didn’t provide the option for graphics alternate text, it was on Microsoft. Now that the feature is there, any inaccessible document, due to lack of alternate text, is on the content creator - especially since Office also provides a built-in accessibility checker.

You’re raising a false dichotomy fallacy. Yes, I agree that it’s hard to have a 100% accessible product. But you can say the same thing about information security. It is IMPOSSIBLE to have a 100% secure information technology product. Every additional “9” increase, e.g., 99.99999%., after a certain point, doubles the cost. However, there are vast resources spent on information security, plenty of government and industry standards, and no one questions their value or the value of getting as close as possible.

No one is paying huge settlements for failing to meet esoteric AAA level standards. They are in trouble for the BASICS, lack of keyboard accessibility, lack of alt-text, lack of captioning, lack of structure/navigation tags, bad-contrast, etc.

dbase, like the PC revolution itself, and airtable (if you’re saying its adoption in enterprises is similar) are examples of niche markets that went viral because of their perceived business value. Paradigm shifts usually comes from OUTSIDE the mainstream, in this case, Enterprise IT. The problems arise when those products then move into enterprise level processes where they introduce information security, reliability, availability, capacity, and accessibility., risks. Businesses have to weigh the value of such innovations while maintaining adequate controls to protect themselves and their stakeholders.

The risk we manage is not just to us financially/legally, but to real people with disabilities who end up being excluded from educational opportunities because someone built and made available, often for free, an “attractive nuisance” that some faculty member, unwitting of the issues, adopts for their course.

I work with A LOT of manufacturers of IT products. Airtable’s response was at the lower end of the curve of responsibility for, and interest in, accessibility. It reminds me of a lot of companies I worked with 10 years ago. If they were more responsive and committed we would probably allow some use of it, assuming that improvements would be forthcoming. But the low level of compliance and their attitude doesn’t let us do that.

Lastly, I recently got a call from a company, that I had pushed on accessibility a few years ago. They called to thank me because their primary market was higher ed, they hadn’t really understood the important of accessibility, and after I pushed them, they started getting more and more demands for accessibility - which they wouldn’t have been prepared to meet if I hadn’t pushed them. They were thanking me for helping their business outcomes. That was cool.

This topic was automatically closed 91 days after the last reply. New replies are no longer allowed.