Hacker News new | past | comments | ask | show | jobs | submit login
Leaving Apple and Google: /e/ first beta (hackernoon.com)
320 points by indidea on Sept 14, 2018 | hide | past | favorite | 118 comments

Hey, first of all, I love to see the custom rom scene still doing its thing. I think we need to keep fighting to keep our phones as open and as 'ours' as possible.

But we, and I mean, a big 'we' as everyone in XDA Developers, Lineage OS developers etc, have as a group a tremendous knowledge in this area.In fact, for the most developed phones, we know almost everything about them. If only everyone could get toghether and do something _better_ than an Android derivative...

I've been thinking for a while in Libhybris. There's a lot of people working on tweaking a library that talks to closed vendor blobs, on which every single one of them is different, to be able to use native libraries directly with files that were compiled for a framework that isn't available on the target system.

What if we could use the Android Framework as an API layer, instead of rewriting everything for every single version of every single library ever shipped with an Android phone? I am talking on stripping down most of an Android AOSP image, just up to the point where the System Server kicks in (just up to a modified Zygote or something like that), and then use the open Android API specifications to actually talk to the hardware through it. Then we could make a standard library which would work with _every single Android phone_

Ubuntu touch, Firefox OS were a failure because they weren't compatible with the existing ecosystem and because developers wouldn't invest in making apps for _yet_ another framework. But Sailfish OS has an Android compatibility layer that lets them run Android apps in their Nemo/Mer/Maemo/Meego hybrid. Chrome OS can also run Android apps.

If you can get those two poins toghether, you can have a platform where most of the available phones can run your OS, and you can also take advantage of the million of applications developed for Android. If you add that to either Ubuntu Touch, Purism's OS, KDE Mobile or any other OS which is still in development, you could have the biggest the Community has had for the last 10 years.

Edit: typos everywhere

> What if we could use the Android Framework as an API layer, instead of rewriting everything for every single version of every single library ever shipped with an Android phone? I am talking on stripping down most of an Android AOSP image, just up to the point where the System Server kicks in (just up to a modified Zygote or something like that), and then use the open Android API specifications to actually talk to the hardware through it.

Sounds like project Treble?

I have vague recollecting of FirefoxOS being like that, where the underlying layer (gonk?) was Android-derived.

They certainly didn't go for an Android-compatible app layer though.

Yes, Gonk is composed of the Android kernel, hardware abstraction layer and some userspace daemons.

With HIDL it should be easier to build a clean abstraction on top of any Android base starting at Oreo.

Sounds like Halium with a dose of Anbox.

Why aren’t companies like Microsoft, Amazon, Facebook not setting aside a few million dollars a year to support efforts like this and the Firefox phone?

Heck, even the carriers would be so much better off if they spent some money on similar projects. They don’t need to spend too much, and if one of them takes off, especially if it’s open source, then it helps reduce the tremendous risks they face since the mobile OS market is currently completely owned by 2 of their competitors.

I assume because it would be competition against any future phone OS from Microsoft, Amazon, or Facebook.

Microsoft is in the phone business too.

Still? I thought they had spun off their mobile hardware division and were instead mostly focusing on mobile software. Would make some sense for them to try to do the same thing they did with PCs and IBM back in the day, since the OS is already open source (mostly).

The rumors are they're still doing one last ditch, Surface Phone due to the growing Surface ecosystem that is making them money and people liking it so far.

However, they're not going to release that until they finish the CShell that they're working on the past several years.

Is Surface making money or just stemming the bleeding?

Microsoft is doing pretty well financially. MSFT stock is up 50% ($75-->$113) YTD.

Sort of. They're in the process of exiting the market.

No new phones being manufactured as of about a year ago, and support is ending at the end of next year.

Yes that's why they said "2". (Google is in the spyware market, hence cannot be taken seriously).

They gave up on mobile years ago. Both hw and sw.

Not entirely true. They shifted focus on mobile to continue making mobile software for other OSes.

We were talking about Windows Phone, not Outlook for iOS or stuff like that.

Which makes it interesting for them to have an OS that won't lock particular software out in a play akin to what MS is trying to to to Valve by using their monopoly, a dogmatic Richard Stallman approach to this would ironically save them from that.

Forked from Lineage but it's unclear what problems are being solved that aren't already solved by Lineage.

Worse, it's forked from an old version of LineageOS that will be deprecated/unsupported once 16.0 is out.

Are they skipping 15 for a bunch of phones? I'm on a 2015 Motorola phone that still gets weekly 14.1 updates, and I kinda wonder if it's just EOL even for Lineage.

> Are they skipping 15 for a bunch of phones?

If your device doesn't already support 15, then it's highly unlikely that it will ever seen an official release with 15, or 16, or anything >14.1.

It's probably EOL.

Generally speaking you still depend on the device manufacturer to ship a compatible vendor partition in order to run a given version of LineageOS. This means that if your device doesn't have an official version of Android 8, then it will never be able to run LineageOS 15.

I wouldn't call it a hard fork, more a downstream derivative. In a similar way, Ubuntu rebases itself on debian unstable every 6 months or so.

That is to say Duval based his product on whichever LineageOS release was 'current' at the time but plans to sync to the next available release when his next development cycle allows (which is Oreo, currently).

Smartphones are heavily tied to services. I think their goal is to create services that respect privacy and integrate them with their fork of Lineage.

In other words, they're making the complete package and they plan to target average people.

Services is the thing /e/ is also trying to solve. They plan to make the complete package. For the tech savvy, you'll be able to install those services on your own server.

“BlissLauncher has original icons“

It looks like a very slight variation of iOS. Not particularly original at all.

I was going to reply and say "original" just means not directly copied, but now that I look back at the screenshot, I would believe anyone telling me that it was of a new iOS version.

My exact thought.

They even have round corners.

This looks very interesting. I'm not sure I understanding all of the forking though - why not just leverage LineageOS, K9 etc. as is just as to leave more time/energy for the other bits rather than having to separate maintain (security patches etc.) the forked bits?

Great initiative, have been looking for an alternative for some time. I had high hopes for Sailfish (https://sailfishos.org/) but for a while it only ran on Sony Xperia, current status not clear to me. Anyone has experience with Sailfish?

Will have a look at /e/ this weekend!

I ran it a long time back on a nexus 4, was pretty good. My biggest problem with it is apps. There's 2 or 3 apps that I really need and these are iOS or Android only

For the love of all that is holy give me back my scroll on your website. https://e.foundation/ I thought we got rid of this scrolljacking nightmare.

Why do people do this? It makes for dreadful UX. Reimplementing scrolling is already a sin; in this case it's even worse. The whole website feels slow, the scrolling is confusing and inconsistent. Same thing with floating navbars and other fancy stuff, don't do it if you're unable to make it perfect.

This post is on a Medium blog.

Life is compromise.

I can't wait for Medium to go the way of blogger, into the halls of internet obscurity.

I'm just worried it will go the way of geocities.

doubtful, nobody will ever look back wistfully on the aesthetic of early Medium pages.

I am guessing the comment was more about how geocities closed down one day and whatever interesting content it contained (alongside uninteresting parts no doubt) disappeared with it.

I really wish someone would have the balls to do something really ass-kicking, like make a PhoneOS that can run all your functions and still give you something special .. it seems to be within grasp in various and sundry vertical embedded markets, but nobody has the druthers to take on Apple.

So really, its not just the OS/distribution/custom-tweaked-environment/game-engine. Its also the hardware.

Like, how can I get the hardware I need to try this out, without going through hoops?

To me that's a bit more of an issue than a broiled and flailed distro-dejour re-incarnate...

Do they audit the apps in their app store to make sure they are safe?

I don't understand what this Android fork has to do with Apple.

The launcher!

I just want a high end phone that supports something like lineageos, has a micro SD slot, and decent battery life. Somehow it's hard to find.

It's getting a bit long in the tooth, but the Samsung S5 has a microSD slot (up to 128Gb), is waterproof, and has decent battery life. What is more, just pop up the back and swap a new battery (US$14 off of Amazon) in. And it has LineageOS 15.1 on it. Again, it is older, but is a high end phone.

Xiaomi Picophone F1. SD845, 4-6GB RAM, micro-sd slot, 4000mAh. Popular enough for treble and lineageOS dev last I've checked... I'm actually using a whyred (RN5) now which is it's cheaper cousin with SD635 and fully flourishing custom ROM scenes.

It looks very interesting (Pocophone, right? Not Picophone? [Even though "Picophone" sounds better to me]). But it's just out, and that too only in India? or at least not in the US, and it doesn't seem to support all of the US bands.

Also, no lineageos support at this point. not surprising, given its newness. But looks like one to keep an eye on!

Then go with Oneplus. 2 years and 2 months and my oneplus 3 with dual sim or 1 sim/ 1 sd card is still working perfectly fine and receiving updates, getting Android P before 2019 (in theory)

Latest one doesn't have an SD slot apparently

Google says the OnePlus 3 doesn't have it either, is that wrong?

Oneplus has never offered an sd card slot

Oneplus X had one. Dual sim and second sim could be used as an SD card.

Any Samsung phone?

Samsungs have crappy battery life and no SD slot. I know because I'm typing this on a Galaxy S6.

> Samsungs have crappy battery life and no SD slot. I know because I'm typing this on a Galaxy S6.

Samsung brought back SD to their flagship line with the S7/Note7 generation. The S6 / Note5 generation is the only one that didn't have SD, as I recall.

Galaxy S8/S9 do have micro SD, battery life still not great but on par with other similar phones I think.

And - referring back to one of the reasons why /e/ was created - Samsung has crappy, horrible privacy practices.

Love /e/ .. waiting for it. Loved eelo better (I know: the name clashed). I guess naming your phone e is one more way to become untraceable by Google (or DDG for that matter). Luckily there is also the 'foundation' part to search for.

When I see things like these, I always think. Oh look, another thing that will eventually get bought out.

I mean, I woulnd't want to depend on one of these things unless the people behind them has shown a enough defiance against the buying power of money.

We really need more than two main players in the the mobile OS game.

"no google no play(store)"

This looks like a great project ;)

The target demo seems to be one that wouldn't care for this product.

“be far more respectful of user’s data privacy”

“the chat application is Telegram”

Is this satire?

I thought Telegram supported user privacy? Please explain.

What if I want to have google services?

While it's awesome you're building a competing infrastructure, it's dead in the water (at least to me) if it doesn't integrate with what I have already.

Its not for you then. The project is about getting away from Google.

Has anyone managed to extract the launcher?

if this work will be great! and ppl are not so concerned about privacy but the google imposition of google apps and services into their phones. that's one of factors ppl use linux and bsd.


> We decided to adopt “/e/” as a textual representation for the symbol of “my data is MY data”

Discussion appears to be here: https://www.indidea.org/gael/blog/leaving-apple-and-google-e...

I appreciate the reason for the name change, but if the goal is to target laypeople, /e/ is not obviously pronounceable or meaningful for my mom or kids/teens.

The other thing is that `/e/`, with the slashes and all, is the exact name of 4chans softcore hentai board.

Let's just be glad they didn't name it /d/, and then use it to run a Squid cluster.

and without the slashes one of many names of a particularly popular party drug

"Hentai, ecstacy, and data ownership? Take my money." - Grandma

To be fair, people call drugs by myriads of names. Context usually helps to disambiguate between LSD and MySQL.

The default MySQL table engine (MyISAM) isn't actually LSD compliant.

MyISAM hasn't been the default since 2009

Call it Evasion OS. I think it's a cool name that has an "e" in it, and it's really about evading the massive data gathering going on.

May as well call it extraterrestrial os because no one on Earth is going to use it

That’s pretty brutal, I’m sure they will have a niche of users.


Not gonna ask. Nope. Not gonna.

Which is why the section in the article continues:

> We understand that this textual representation is not ideal for talking about the project and searching about it.

> We have decided to hold off on change to the name for the beta release. Along with the v1 stable version, we will introduce a new name for the mobile system that will be easy to remember and to adopt.

So: They are aware and have a plan.

I just assumed it was pronounced “slashy-slash”...

That is awesome. I like this name/pronunciation!

And it's very difficult to search for

This should be higher in branding concerns today

If you're not googleable, you're going to have a hard time

I actually miss the days of bastardized short words as product names. Yeh some of them got a bit ridiculous, but they were unique and easily searchable. Flickr, Google, etc. People were creating brands that stood out, were memorable, and they owned their entire results pages.

Now we're using common nouns as names for everything and it has completely broken the ability to search for both the product AND the common nouns.

Totally agree, and I think it’s extremely important to have better branding if you’re actually trying to reach those user groups

This kind of project will never get any kind of attention by laypeople, so it's good that they know that and aren't even trying. Less wasted energy :-)

It's similar to //e which is how Apple IIe was stylized. If it's not going to be a trademark issue, it seems like a clever branding hack to me. Anyone who grew up in the 80s & early 90s or earlier might be subtly influenced by it.

I think /e/ cute and that plenty of brands have succeeded despite being unrecognizable and unsearchable.

The meaningful symbol seems related to AirBnb and their bélo [1], which I think was largely successful.

1: https://www.youtube.com/watch?v=nMITXMrrVQU

IIRC It was stylized IIe and //c

I thought the ASCII art version was ][e

"slash e slash"

"slashy slash"

"Slashy McSlash Face"

> /e/ is not obviously pronounceable

On the contrary, it's a scientific notation for a specific pronunciation of the letter e.

/e/ represents in linguistics an unrounded closed-mid front vowel.


Keyword is obviously

A naming that encodes its own pronunciation is genius.

'Linux' was non obvious to the point that Linus issued his own voice recording. 'gif' vs 'jif' for the image format, etc.

The name will be the largest barrier to entry.

It is not easy to say, remember, or look for.

This was dead before it even got off the ground, which is unfortunate because the OS looks nice.

I tried typing the name /e/ into my address bar, and literally could not get my browser to trigger a search. It kept assuming it was a path to a file!


    ? /e/

Isn't this the point?

Calling it "eOS" could be a viable name for lay people. Officially it can still be /e/.

eOS has trademark issues. Same reason that elementaryOS doesn't use it and actively discourages users from doing so. Though... users call it that anyways.

OP is right - they should really fix their name to address all these issues: (1) Search engine-ability. Zero people are going to be able to discover this. (2) Unique domain name. e.com/io/whatever is already taken (so is eos.com). (3) Not trademarked.

This is kind of already taken by elementary os though.

How about something phonetic like Slashy Slash (a la Marky Mark)?

Looks like it's already taken by a number of things:


He wrote that they will change the name when they release the first version by 2019

He should probably change it before then

Judging by the name I thought this was a 4chan board.

At least it's not /r/os

I understand and welcome more alternatives to Lineage but what's wrong with their user interface?

If you want to make a better Lineage, I think having built in default app stores like Apkpure, encrypted system proxies, isolating apps from each other in better ways... Aren't those more worthy privacy oriented things to work on?

None of those will get you media attention, a successful kickstarter and funding, though.

Boring. I'd like to see a smartphone that ditches the "app" model (where your phone is an appendage of other people's brands) to a "personal assistant" (where your phone works for you.)

I would like the OS to have 1 messaging app, 1 email app, 1 calendar app etc. Then underneath, each provider has their own interface to their backend. Then we can integrate everything into one system.

Now this idea still needs a lot of work. But, IMO we need to get away from the app model onto some new paradigm. Present us with our information /data directly without having to open apps. Think android widgets on steroids. Windows phone actually had some of this, with letting you create tiles that where contacts or other info.

I like that idea, but I think the real world implementation of that would be one huge monolith app (still owned by one company) like WeChat. Maybe if Google made an app which integrated all their services :P

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact