Hacker News Reader

Comment by jeswin | Hacker News Reader

Having spent a couple of years building apps for KaiOS (which is really Firefox OS), I can say that it's probably the easiest development platform out there - mostly pure web; with some proprietary extensions.

I wish Firefox OS hadn't thrown in the towel so early. These devices are selling millions of units every month.

View on HN

firefox os had so much potential, now with pwa's being the next big thing, it seem like firefox os could be a game changer. Too bad they killed it.

what is pwa ?

Progressive Web Apps

Progressive Web App. A web app that's supposed to be indistinguishable from the native one, built using web technologies (so that you don't have to maintain multiple code bases).

When Firefox OS was around, they were still pretty niche.

Firefox OS was doomed to fail. They created an operating system which used the slowest rendering and JS engine there was and put it in the worst hardware they could find.

These phones run on the same rendering and JS engine...

"KaiOS apps are based on web technologies–HTML, CSS, and JavaScript–and are run by Gecko runtime."

https://developer.kaiostech.com/

Except that Firefox is phasing out Gecko and replacing it with Servo (we're not yet there but some components have been replaced already - this is what the Firefox Quantum thing was all about). So this KaiOS thing is basically picking up an abandoned project...

Servo is nowhere near production-ready and there's no reason to assume that it will be anytime soon.

Just consider that Mozilla is pretty much working full-pelt on Gecko and merely keeping up with Google. They (as well as Google) are far away from implementing all currently specified web-standards, of which more get specified all the time.

So, in order to get Servo to the level of current browser engines, they would have to have an even higher developmemt velocity, while not really being able to stop developing Gecko in the meantime either.

Maybe if the majority of components someday are shared between Gecko and Servo, they might do the final step and switch out the core completely, but even that is still far away.

Servo is specifically a research project. To explore what could be done, if one were to do things right. That they were able to isolate and share components, that even came as a surprise to Mozilla.

Also, Servo isn't a general-purpose browser project anymore at Mozilla. It's part of the "Mixed Reality" department:

https://blog.servo.org/2018/03/09/servo-and-mixed-reality/

Servo, while a research project, is also a place to build the fastest possible implementation of a browser (w/o full backwards compatibility), then crib the best performing parts and move that into mainline Firefox (what Project Quantum is, a metered replacement of Gecko).

Swapping out the jet engine of Firefox in flight (Gecko) was likely to be a very bumpy, messy road, thus Mozilla has chosen to break the problem down into manageable chunks by having a parallel team build Servo and push the bleeding edge of performance & features, while having another team break the new code (Servo) and the old code (Gecko) apart into the separate , interchangeable pieces, then clean up standards compliance in the new Servo module and prep it to be replaced.

Essentially, its the Cathedral vs the Bazaar all over again. Microsoft Edge was a ground up rewrite of most of the browser (the Cathedral model of software dev) along with Android (built behind closed doors at Google, then code dumped right before a new major release). Mozilla didn't like how this model could easily backfire, choosing a more metered approach by developing smaller, swappable components in the open.

https://wiki.mozilla.org/Quantum

Gecko is the rendering engine. The JS engine is called SpiderMonkey.

I'd be interested in this on a modern smartphone platform with a touchscreen and virtual keyboard just to cut Google out of my life.

> just to cut Google out of my life.

Apple?

Out of the frying pan into the fire...

So it looks, but, realistically is the only option for phones. And it is not an adtech company, you pay for hardware and software.

Why not run Lineage OS then?

I dunno, switching from a phone with a pre-installed Google maintained OS to a custom rom of a Google maintained OS isn't exactally fully cutting Google out.

AOSP and its derivatives may be maintained by Google but they don't call home without the proprietary Google apps. Just leave those off and use something like F-Droid to get apps and you're Google-free. None of the Android devices I use (phones and tablets) has any Google-proprietary code installed, all of them work fine without such.

Lineage is Android sans Google. It's exactly what you want if you want to cut Google out.

You could run https://e.foundation/

Wait for Pine Phone for $150 [0] or preorder Librem 5 for $649 [1]. However expect beta experience either way.

Also remember that you probably don't want to entirely cut Google out of your life, as for example Linux have Google salaried developers. But that's, I imagine, not what you meant.

[0] https://forum.pine64.org/showthread.php?tid=7339 - no canonical site yet

[1] https://puri.sm/products/librem-5/

Why wait for a $150 phone or preorder a $649 phone while more or less the same thing is available on a $7 phone in India? I think that $7 phone should go on sale in other countries as well, either in its current guise or - probably at a higher price - without the Google bits. It would certainly be an interesting device to use in circumstances where pricey and vulnerable smart phones don't make sense - e.g. when crossing unfriendly borders, when working outside in the dirt where that smart phone always runs the risk of being damaged or lost, etc.

Google invested in it not for it to be available without them. So I say it's not gonna happen even with higher price. They would also not like a suggestion that the same phone without their software is somehow more valuable. Also I feel that the GP asked for less hypothetical options, not more.

I would certainly pay more to avoid Google.

"Linux-based 4G phones with Google Assistant " i was ok with the first phrase, the second was like grain alcohol spiking a fine bourbon.

The reason i would buy a linux phone would be to avoid having a MITM burned onto the rom of my phone. The device presented in this article doesnt provide that nessecity.

I think the law of averages applies, not knowing how much google and others payout to burn kruft into peoples product, im going to guess that the increased global saturation level of google ASSistant is worth it for google and a good revenue for the OEM. ill bet the price is subsidied to 7$, i would pay 70$ for this phone, if it had no google ass in it.

They needed the resources to integrate pocket into firefox

It's kind of funny that PWA was Jobs original vision for iOS apps with HTML5, before they came up with the walled garden cash cow approach.

> “The full Safari engine is inside of iPhone. And so, you can write amazing Web 2.0 and Ajax apps that look exactly and behave exactly like apps on the iPhone. And these apps can integrate perfectly with iPhone services. And guess what? There’s no SDK that you need! You’ve got everything you need if you know how to write apps using the most modern web standards to write amazing apps for the iPhone today. So developers, we think we’ve got a very sweet story for you. You can begin building your iPhone apps today.” - Steve Jobs/iPhone announcement Jan 2007 at macworld

https://medium.com/datadriveninvestor/2019-the-year-of-progr...

What's old, is new again, 12 years later. It's how it always should have been. And guess what? There's no SDK that you need!! ;)

An idea that was already being explored on Symbian with their WebRuntime actually.

https://web.archive.org/web/20100603124720/http://www.forum....

I feel like the whole Symbian ecosystem disappeared, it's a shame nobody is preserving any of the software, that was 10+ years of tech that is basically gone.

Quite true.

All what is left seems to be Web Archive and these sites.

http://www.allaboutsymbian.com/

It's not like this went away, though. You could always, and still can, write iOS apps this way. Safari even has the "add to home screen" feature which lets you run these apps without the Safari chrome around them.

Safari/Webkit still lags behind the other browsers, and finally started adding the basic features for PWAs last year.

It's nice that a lot has been implemented, albeit slowly. Though some may still think about Apple incentives.

I.e.: http://nolanlawson.com/2015/06/30/safari-is-the-new-ie/ https://news.ycombinator.com/item?id=19586219

PWAs make sense today, but back then, they just came too early in the game. The iPhone only had EDGE (2G) networking, it was painfully slow both in terms of throughput and latency. Each HTTP request would take ~300-500ms under optimal cell coverage.

Additionally, unlike modern PWAs, web-apps back then had persistence for data only, so you had to reload the entire application over the network at each launch. People back then loved the switch to native apps because they were extremely snappier.

Worked well for i-mode in 1999 - of course that did do some magic on the protocol level.

https://en.m.wikipedia.org/wiki/I-mode

that doesn't make any sense. All the limitations you mention also applied to apps.

the whole problem is that apple shipped native apps with tons of skeumorphism and animations, so all js frameworks for webapps spend all their efforts to simulate "slide animation on page change" instead of working on hiding the ajax latency or using local storage better.

and in the end, everyone just saw $$$ in using the closed apps from the store.

> PWAs make sense today, but back then, they just came too early in the game.

Seemed to work well enough on webOS for the Palm Pre.

I don't know about its potential but what was there was not a serious challenger to the existing platforms.

I was working for a very large app (tens of millions of daily users) at the time when firefox was trying to convince companies to port their apps to it.

We were big enough that Firefox sent their advocates to our offices and tried to convince us to port our app. Since what we were offering was a service, we were naturally open to have it on as many platforms as possible. I think that there might have been a monetary incentive as well but I might be misremembering there (I was not part of that negotiation).

After a week with their sdk and help from their devs, we threw the towel.

Publishing our product on Fos would have been impossible while keeping the app quality on par with iOS and Android and there were lots of missing/buggy APIs.

That was our last discussion with Firefox. For as long as there have been phones, there have been web apps targeting them, but there is a reason why we are still building native apps.

Are PWAs the next big thing? I thought the hype for PWAs had for the most part fizzled due to lack of user interest.

It’s not the users that are pushing native apps on every website. It’s not even the developers. It’s the media and add companies that realise that they can get away with things in a native app that a browser would never allow.

I hope it's got a lot more polished, I got one of the first firefox phones and it was pretty unusable in most areas.

I pretty quickly got an Android phone instead.

Like the old Palm Pre?

WebOS was ahead of its time.

Symbian as well.

Many of the supposed iOS "innovations" were already there.

Is, not "was." My LG television runs WebOS.

...as a conduit to barf ads at you. I wish I could have gotten my TV without ask the parts that forced me to install a PiHole just to not get ads overlaid onto my menus. Any good parts of webOS are now just incidental.

I just banned the TV’s MAC address from my network. IMO there’s no reason for it to connect to the outside world.

Good solution if you can do it, but my girlfriend likes watching YouTube and Sling through it. :-/

The WebOS that runs in your television is very unlike the WebOS that was.

That ecosystem, interface, and community no longer exist.

Also your comment doesn’t really make sense.

Something that currently exists can still have been ahead of its time. My comment isn’t an indication it doesn’t exist.

> KaiOS (which is really Firefox OS)

Thats news to me. I read that KaiOS was built from the ground up & not derived from anything.

> Firefox OS hadn't thrown in the towel so early

That was a rather quite event and I was surprised to see no much discussion around it. An OS by Mozilla can be more trusted than an OS by Google. Why did Mozilla kill the project?

> Thats news to me. I read that KaiOS was built from the ground up & not derived from anything.

Where did you read that? Read the first sentence of: https://en.wikipedia.org/wiki/KaiOS (all with good citations, e.g., https://support.kaiostech.com/support/solutions/articles/350... )

Wow. I didn't check the wikipedia entry for KaiOS earlier.

But the weird thing is that I recently read in some magazine (I am unable to remember the name) about Google investing in KaiOS which was a Brazilian company and the whole article was mentioning how Silicon Valley is unable to understand its customers as they live in a bubble and how its the companies in developing countries that truly understand the needs of the customer and deliver. Its mentioned that KaiOS is gunning at Android for #1 in India and Google's investment was a sign of acceptance and a hint that the new version of Android will get shelved soon. It even referred to India based Zoho as the true challenger to Google Apps & Office 360. Also how wechat is miles ahead of Whatsapp & likes and how in a sense it challenges the whole Silicon Valley in terms of all the features it has. Basically it was urging the readers to not assume that silicon valley products as superior and try others too.

I hope Mozilla reconsiders and starts up Firefox OS again. I prefer it over a company driven product like KaiOS.

P.S. Just noticed that the last name of the KaiOS's CEO is "Codeville".

Is there much difference between development of a KaiOS app, and a ChromiumOS app?

Was Firefox OS targetting this kind of phone though? I thought it was trying to be a touch-screen smartphone OS.

Wasn't really, although their focus was on low cost phones too. Credit to KaiOS for recognizing the potential of non-touch devices.

However, the effort required to enable a non-touch UI would have been a small fraction of the overall FirefoxOS effort. Something Mozilla could have responded to, if the team was intact.

The non-touch concept was under development when firefox os was cancelled.

Yup. FFOS actually convinced me that even high-end devices built on open web technologies could work quite well enough for today's usecases. I'd buy one today if I could - not a low end device but a Snapdragon 8xx device.

Web is good enough for 90% of apps, and a bare metal SDK would be needed for games.

PWAs did it for me.

Even WebGL 2.0 alongside WebAssembly will be good enough for Flash like games (Google Chrome team will give a talk about WebGPU at IO).

Reply on news.ycombinator.com