davidascher

So few people use Metapost, it's undervalued! Feature request: add the output to the github repo so we can see the result?

See https://voting.works

Reminds me of this ancient Firefox bug about the performance cost of the throbber: https://bugzilla.mozilla.org/show_bug.cgi?id=437829

HTTP only sign up page? really? in 2017?

As others have said, management (in my experience) never has the full picture, it has its own perspective.

I'm curious about the breach of trust you feel -- surely investors in a company don't relinquish their values or place unconditional trust in the company? The Kapors are not a PE firm, and their ethical stands have been known for decades.

You can't on the one hand benefit from the brand value of investors like Mitch & Freida and not also expect such stakeholders to also expect a certain caliber of behavior by the company.

I think investors in a startup are forgiving of early missteps. Uber is way, way past that stage.

CZI is focused on disease. A focus on biomed research is reasonable...

Out of curiosity, have you looked at Mozilla's Pontoon and if yes, can you compare/contrast the goals of Parrot w/ those of Pontoon? https://github.com/mozilla/pontoon

Yup. And such a chef probably wouldn't publicly berate people for eating meat, or expect being called out if they did.

Somewhat funny to have an FB employee complaining about overly centralized systems.

This was done in partnership with the Internet Archive folks -- they _want_ the traffic, in part because this helps people know about the Archive, but mostly because this is a public benefit feature, which is what non-profits strive to provide.

As to the amount of traffic it generates, that's one reason why we're doing this through Test Pilot -- we can dial that up or down.

Good questions. At this point, we're trying to identify user value with prototypes. We're a long way away from defining things like the software update model on as-yet-unspecified devices.

Sounds interesting. Do you have any references for approaches and possible benefits that we could look at?

MDN is powered by Corp staff but also a large # of awesome volunteer contributors.

Check out the code here: https://github.com/mozilla/donate.mozilla.org

TLDR: stripe for donations, custom software for the forms.

Correct. (note: I believe none of our users have asked that ;-), but I appreciate that HN cares about that.

Hey all --

David Ascher here, from Mozilla.

Thanks for the interest. We'll put up a technical write-up of Webmaker soon, we've been busy getting the app ready.

As to some of the questions already asked:

This is an open source project (both the Android wrapper, the web app and the backends powering the APIs). Repos are here:

* https://github.com/mozilla/webmaker-android

* https://github.com/mozilla/webmaker-core

* https://github.com/mozilla/api.webmaker.org

When it comes to data privacy, remember that Webmaker is not a social network. It is an easy-to-use publishing platform for people whose only internet access may be from a low-end Android phone. We don't ask for or store real names to create user accounts.

All projects on the platform are publicly accessible/remixable under a CC-license (it's a free, no-ad app -- all we ask from our users is that they contribute to the commons). We currently store all user data on Amazon RDS.

You may be interested in some of the user research we did as part of this project as well:

* http://mzl.la/india

* http://mzl.la/bangladesh

* http://mzl.la/kenya

* http://mzl.la/research

Hi there. David Ascher here, from Mozilla.

* No Real Names requirement. * I think we have a single account per email address, but I don't think that's what you mean. Also, you can likely use +whatever on your email addresses if you really want to.

As to centralization -- I'm as worried about centralization of the net as Brewster Kahle, but that's as yet an unsolved problem, and not the one we're tackling with Webmaker. We're (just?) trying to make an app that makes it easy for people new to the internet to see that they can create web content, not just consume it, even if all they have is a low-end android phone and low literacy. The app storage architecture is centralized today, but I'd be happy to explore ways to federate it, once we get to the point where it we've confirmed the app is attractive to its main audience.

TLDR: Sorry, this isn't the droid you're looking for =)

I love the spirit of this -- A real "public service". thanks.

The importer data is way more provocative to me than the exporter data, which met my expectations.

Ugh. Thanks, will get folks to look at what happened.

so that non-devs can do edits via editing, dropbox syncs, push using web UI, and have the magic happen w/o needing devtools everywhere.

This is one path: http://www.mozilla.org/careers

Partial proof: https://github.com/mozilla/repositories https://github.com/mozilla-b2g/repositories

(that's 30 _pages_ of github repos, not including the stuff we do that's not on github: http://mxr.mozilla.org/)

Crawling is certainly a complementary data collection strategy, but it's harder to avoid IP-based "filter bubble" effects w/out also deploying something akin to a bot. The hope is that by using real people using real browsers we'll collect data that reflects actual-behavior-in-the-wild.

You're right that poisoning is a potential problem if/when the data ends up useful enough to warrant poisoning.

Thanks, great set of feedback. A bunch of these need to find their ways into our issues list!

True, but most of all we value working in the open so that we can both do user testing and find people who might want to contribute. In my book our mistake was not to prioritze issues 306 and 386 until we showed up on HN. Oh well!

Filed: https://github.com/mozilla/appmaker/issues/389

I love this comment. "Get it working". Roger wilco!

Ha. It's intended to look like a FirefoxOS phone if anything =)

Indeed, we're inspired by AppInventor. We're aiming at higher level components and a more simplistic model.

Indeed, we have a weird intermittent failure on component loading that we haven't figured out yet. A few reloads tends to work, as a temporary workaround.