cmaury

Are you able to pull upcoming titles? All I want is a weekly/monthly list of books by authors I've ready which are coming out, and I've not been able to find it or to build it.

I like it, though I'm sure we'll end up being stuck with "vibe ware"

Thanks for sharing. This is exactly the type of utility that vibecoding is for. It takes 5 secons to ask GPT to write a scripr to do this tailored to your specific use case. It's way faster than trying to get someone elses repo up and running.

Riffs need a downvote button. I'm not sure what the team's training/eval approach is, but you need a user signal for poor quality results, so you can improve over time.

ex: https://www.riffusion.com/riffs/2a42259b-1557-4913-bc04-5f0a...

Amazing progress since the original announcement.

But do you have a YouTube channel? (jk)

Are you working remotely with this set up?

Hey, I'm a legally blind engineering manager (stargardt's macular degeneration). The accessibility community was hugely welcoming when I started to lose my vision at about the same age you are. I'm happy to talk through my experience and be helpful in anyway I can. chrmaury at gmail.

How do you see those two as being different?

In my mind, products that make use of ML are using ML services as the back end. An example would be the recent wave of bot companies: many are not rolling their own NLU system but rather leveraging services like wit.ai or Microsoft's LUIS.

While Google may be the leader in Machine Learning tools, they are way behind when it comes to Machine Learning services.

They've only just now released a Natural Language Understanding service in beta, and it is more limited compared to other NLU services from Microsoft/others.

While tooling is important, the market for low level tools is much smaller than for the services built using those tools. The vast majority of businesses who could benefit from Machine learning don't have the expertise to run RNNs using Tensor Flow, but do have engineers who can integrate API's that leverage trained classifiers.

You can check our SDK out at developer.conversantlabs.com. It's currently in a developer preview. Send me an email at chris@conversantlabs.com. It would be great to talk more. If our conversation in this thread is any indication, I think we'll have a pretty good discussion :)

This really wasn't my intent. I sincerely apologize for communicating otherwise.

I identify as blind or low-viz and wouldn't want to take away from others who also identify as such or as disabled.

That line was meant to be a response to the fears that it seemed like my family and friends had about my diagnosis. Fears, that I wouldn't be able to remain independent or be successful. You know how loved ones generally fear the worst possible scenario. Anyway, wording it this was was terrible, and I'm sorry. I'll see if I can make an edit and rephrase things.

I definitely agree with you that the last thing I would want readers to take away from this article, is that they don't have to worry about accessibility or universal design. Until we have better tools, we should be providing the best possible experience with the tools we do have.

I also agree that we shouldn't be creating separate experiences for the blind. I think it's generally acknowledged that they end up being worse than a combined interface, never getting the resources or new features that experiences for sighted users get.

Where we seem to disagree is on the roll that screen readers play in limiting the usability of technology. On the one hand they are amazing because they provide access to technology that would otherwise not exist. On the other hand, by virtue of the way they function-mapping a 2 dimensional visual experience into a one dimensional stream of audio-using a screen reader can only be so efficient.

This lack of usability puts access to technology beyond the reach of many who are less tech savvy than you or I, and given that the vast majority of people losing their vision in the US are the elderly, there are a lot of people who fall into that category. What's worse, the rate of vision loss is set to double as baby-boomer's age out.

I totally agree that the medical-model of Accessibility sucks, but I think Screen Readers fall into that category. They seek to adapt an experience designed for others to the needs of the disabled. Conversational interfaces have the potential to create a consumer quality experience, that by it's very nature is accessible (at least to the blind). And accessible by default is the best possible outcome.

There are a lot of professional developers who are blind. They use accessible IDEs paired with a Screen reader. Here is a good description: http://stackoverflow.com/a/453758/319013

In the future, I think there is a lot of potential for pairing a Conversational UI and a natural language programming...language. Eve looks really promising, and could change the way that everyone writes software, and not just the blind: http://eve-lang.com/

Here is a demo: https://youtu.be/VZQoAKJPbh8?t=46m52s

Sorry for the let down. I feel her pain. We're working on it, and should have something for her soon. Email is definitely high on our list of tasks that need a conversational layer.

You caught me :). It was great meeting you daughter! She seems to be doing great and taking things on with the right attitude.

Feel free to email me anytime, if you ever need anything: Chris@Conversantlabs.com. Happy to talk through things in a more private setting than the comments section of HN.

Absolutely right.

We refer to this as the Scarlett Johanssonn problem. Any system that is smart enough to understand what we mean by the Scarlett Johanssonn problem, and the reference to Artificially Intelligent virtual assistants, would itself need to be a strong AI.

We've already solved the problem of translating human to machine code for graphical user interfaces. It's the desktop with the keyboard and mouse. It's the app and the touchscreen.

You're exactly right that the question now is how can we do the same translation for voice-based systems.

So far the approach has been Virtual assistants, which is really hard, see again the Scarlett Johanssonn problem.

We've been working to find the right design/interaction metaphor, and the app metaphor seems to work really well.

Not only is the search space limited, like you mentioned, but users have a more intuitive since of how to interact with the app. When you open a shopping app, you know that you can likely search for something to buy ("I'm looking for a new pair of shoes"). When you interact with a general virtual assistant, that interaction is less clear.

The app model also works for scaling conversational interaction beyond a single context. Individual apps can support their own language for interaction which best suits their function. This of course necessitates design standards, just like we have for GUIs.

The limitations for fully voice-based applications aren't technical at this point. They are design problems with known solutions. The break out for voice will happen when the right form factor comes about with the right UI design.

This has been commercialized. I'm not sure if it's based on the same research, but you can find it here: www.theaudeo.com (site currently down, but it was up a couple of weeks ago).

"Jobs that pay enough for them to afford the internet"

A strong argument for getting the third world online is to help create more jobs. This sounds like a corporate sound bite, but it's a complex issue where job opportunities are a few degrees removed from providing broad access to the internet in and of itself.

In a lot of developing countries the only jobs that are available are supporting existing industries that are heavily dominated by mega-corporations that are much more focused on resource extraction/profits than they are on developing human capital: Clothing industry in India, Tourism in Mexico/Caribbean, Mining Natural Resources everywhere.

While much of the initial profits and use cases for bring a country online are frivolous and tend to mirror 1st world consumption patterns, the internet access is critical infrastructure for supporting a more decentralized economy.

As adoption of the internet, and telecommunications infrastructure more generally take hold, economic opportunities that didn't exist before will start to emerge: Mobile payments in Kenya, Bitcoin merchants in Chile, or even the selling US produced content in Cuba (the "paquete").

Access to decentralized communication networks enables people in developing countries to participate in the economy on more favorable terms and enables people to work and start businesses without having to work through existing gatekeepers (I think this is why people have responded so negatively to the internet.org model).

I would love to see some studies on internet or network driven economic development, maybe as an alternative to Import Substitution.

Justified paragraphs are mostly useful for text laid out in columns (i.e. Newspapers and magazines). This way when you are reading the second and subsequent columns, the eye can easily find the start of the line/column.

Most blogs are single column. Justifying the text will create variations in the space between words which makes it harder to read with little benefit other than an overall aesthetic that may or may not be more pleasing to look at as a whole.

Another cost effective option is NVDA on PC. It's free and open source.

While this is true, it turns out the lack of buttons isn't as limiting as you might think. You may not have haptic feedback, but you still have the ability to explore the screen using your spacial/tactile senses. By touching the screen and dragging your finger around you can explore what is displayed on the screen. The phone will read aloud the label of what ever item your finger is touching.

For an much more in-depth and eloquent discussion of touch screens and accessibility, you should check out this paper, slide rule which is the research that the VoiceOver design is based on.

http://openexhibits.org/paper/slide-rule-making-mobile-touch...

Unfortunately there isn't a good solution for this, especially on the desktop.

You may have a much better time with an iphone/ipad/ipod touch. The interface is much simpler than a desktop, and VoiceOver is much easier to use than JAWS.

Apps that can help (there isn't one catch all) Apple mail is pretty simple, and fully VO compatible Voice Dream Reader simple text to speech app that syncs with pocket, evernote, dropbox, and most importantly bookshare. a service for the blind with pretty much every book published on it.

Web pages are hard to navigate, but maybe some combination of pocket and voice dream reader will do the trick.

I'm going blind and this is how I perform the tasks you're talking about.

Unfortunately, if these solutions are beyond the means of your mom, there aren't many other options for the tasks you are talking about,

I'm working on an SDK that allows developers to add a conversational interface to mobile apps. This type of siri-like interface turns out to be much easier for seniors to learn than traditional assistive technology. You can check it out at conversantlabs.com

Feel free to message me if you want to talk more about helping your mom out.

I'd also add that creating a cash incentive enables you to encourage an open sourced jailbreak, which is doubly important in this case since given all the opportunities for a jailbreak to introduce malicious code

You're absolutely right that changing motivations from intrinsic to an extrinsic can have negative affects. I'm pretty sure there is a study that people paid a dollar to complete a crossword or some other word puzzle did worse than people who were asked to do it for no pay.

However, I think there are cases where extrinsic motivation can be beneficial. This campaign provides a good example.

Those with the programming skills necessary to jailbreak the iPhone may not be as intrinsically motivated to do so as the OS matures, and so they stop dedicated time. But these engineers are by no means the only people who might benefit from jailbreaking.

The disabled community is one such group. They, in general, may not have the skills necessary to write the software necessary to jailbreak their devices, but could benefit greatly if such software existed. Offering a bounty is a way for them to cross that skill gap.

These campaigns aren't meant for the benefit of software developers. They are for the benefit of those who have great ideas, but lack the skills to fulfill them.

Extrinsic motivation may not be the ideal condition, but that does not mean it is something that should be avoided.

Worrying about updates is a fact of everyday life for people dependent on accessibility features, on the OS as well as the APP level. Updates to apps often include changes to UI, which completely change the way blind users interact with the App via VoiceOver. This means that with every update there is a risk of the app no longer being accessible or having to relearn the entire interface.

It's true that Apple's accessibility features are miles better than Android's, but they aren't perfect. Jailbreaking allows people to add the functionality they need that isn't currently supported.

A great example of this is f.lux which is an amazingly useful app for people with low-vision, but is not supported on non-jailbroken devices.

You're absolutely right. The accessibility features are lightyears better on iOS. This means that far more people with visual impairments are using iOS and thus more apps are being developed for the blind on iOS than other platforms. It's a pretty virtuous cycle :)

How stressful/demanding is your day job? When I was working 12 hour days at a start up, the only thing I was able to do when I got home was surf reddit and go to bed. I was so drained; physically and mentally exhausted.

It wasn't until I left that I was able to put more energy into the things that I enjoyed outside of work: coding, commenting, creating shit, etc.

I dk, it might be worth it to consider doing something else during the day.

Fluid width sites are extremely helpful for accessibility. I have low vision, and so I have to zoom in on most websites. Fluid width allows me to navigate without having to scroll sideways.

I meant an Torrent Index, such as Piratebay

I thought you could limit searches to .onion domains. Oh, well. Thanks for the offer to help. There are a few ideas for site that I would want to see on the Tor network, but I don't think they would have the draw enough traffic to get people running TOR on the regular.

Though writing this gave me an idea: is there a .onion Torrent index?

edit: fixed Tor capitalization

There are already multiple search engines for Tor, including DuckDuckGo.

The problem with Hidden services/ TOR isn't search. It's the fact there aren't enough legitimate/trusted websites that appeal to users outside the hardcore privacy/security crowd.

edit: fixed Tor capitalization.