/e/ Initial Thoughts

In which I try very hard to not wax philosophical on their strategy or the state of Android or talk about how weird the name is and instead just talk about the beta release itself.

My experience is largely from v0.1-201809161427 on a HTC One m8

Observant readers may notice that this commentary is posting several weeks after that build version. C’est la vie.

What is /e/?

/e/ logo

Gaël Duval, founder of Mandrak GNU/Linux ran a crowdfunding campaign to fund the initial development of a mobile OS called /e/ earlier this year. On the surface /e/ is an Android distribution or ROM, the bulk of which is forked from LineageOS. However, Android ROMs are usually shipped somewhat incomplete. Any short time exposed to the LineageOS community or the XDA community will quickly reveal the fact that most Android users running an alternate ROM or distribution are installing google services ontop of it. Even users of microG (A FLOSS clone of some google services) are often trying to get some Google applications installed ontop of microG.

Only a few ROMs or distributions are attempting to be a complete experience on their own. It is and likely will continue to be easy in the short term to dismiss /e/ as a just another ROM; in theory the rubber will meet the road later as they provide more and mature alternatives to google’s services.

Long term the /e/ foundation wants to provide reflashed or OEM phones with /e/ to end users, providing users with an easy to use privacy respecting and floss mobile experience coupled with a user privacy education campaign.

What makes it different today?

One of the biggest immediate differences is the bliss launcher. It’s honestly one of the biggest puzzles to me within the entire /e/ project so far. It’s not a bad launcher in fact ive grown quite fond of it but it is clearly more than inspired by iOS. Application icons are clearly done in iOS style; there is no android style app drawer. App icons even are rearranged by long pressing on them and watching them shake; and while I have never owned an iOS device this is so quintessentially Apple that im even aware how shamelessly this is all lifted.

Gaël Duval had put a lot of emphasis of Android needing a usability overhaul for this to succeed and really promoted the launcher from day one of the kickstarter. I must confess I would have enjoyed something more original feeling; and the amount of exposure bliss launcher has had on Gaël Duval’s Patreon makes me suspect it’s received a disproportionate amount of development time; however its a point im a little apathetic on as well. Obviously many find the iOS UI experience superior to Android which is in part why iOS retains so much marketshare. Bliss also still has a lot of growing to do and hopefully will distinguish itself more in the future.

The Bliss launcher from /e/

The second way /e/ distinguishes itself today is via the preloaded apps. Eelo provides bundled apps that exceed the normal AOSP apps in baseline android and are meant to be good replacements for the proprietary google bundled offerings. Telegram and Signal come preloaded as replacements for WhatsApp/Facebook Messenger/Google Hangouts/SMS. A fork of K9-Mail whose UI has been redone in material design style is supplied as a replacement for Gmail/Mail. Search widgets point to a /e/ foundation hosted and branded Searx instance.The Google Maps replacement being the only one which is non FLOSS. Gaël Duval is obviously uncomfortable that the maps replacement is non floss for now. He has gone into some light details on his reasoning and specifics on the propriety app’s privacy policy on the /e/ wiki here.

The third way /e/ distinguishes itself today is with cloud services. One ways the /e/ foundation hopes to generate sustainable revenue long term is by selling storage for an optional /e/ account. /e/ hosts email and storage for users much like a google account provides access to gmail and google drive. The goal is to allow users to self host or to rely on the foundation’s resources. For the time being these are largely existing floss projects such as NextCloud but with /e/ branding though some care has been taken to unify and streamline the setup of such services. When one sets up their account via the phone UI, email, caldav, carddav and file syncing are configured simultaneously and seamlessly; the included notes and tasks applications likewise automatically are configured as well. Each service can also be enabled or disabled independently of each other from the /e/ account settings screen.

The rough bits

Bliss launcher is coming out of the gate fairly usable but with virtually no possibility to configure it. It’s search widget can’t be pointed at another search engine yet, (But the Searx instance can be configured to show results from specific engines.) There is no choice of what widgets to display or how to disable them.

While some care has been made to unify various disparate apps within /e/ others are not. I for example would love to see the included weather app and weather widget draw from the same pool of settings so that configuring one configured the other. (And tapping the weather widget launched the other.)

File synchronization extended to app setting synchronization was interesting but aggressive enough I had to turn it off. There was no way to exclude apps or folders from the synchronization process and I quickly found podcast episodes I downloaded on my phone were being synced up to the /e/ NextCloud instance.

No app repository yet.

Some preloaded app choices were odd. There is a lot of use case overlap between Telegram and Signal. A lot of Gaël Duval’s public comments on the app choices seem to make it clear he prefers Telegram as the “WhatsApp” clone and Signal for SMS. Packaging Telegram and Silence would have made more sense to me personally; also with the emphasis on self hosting and or /e/ foundation hosting makes the absence of XMPP/Jabber or Matrix seem odd.

The good bits

Bliss launcher has grown on me in the short time ive been exposed to it, and im looking forward to it maturing.

The fact that Android doesn’t come with caldav/carddav out of the box was always a kind of travesty, I was pleased that /e/ comes preloaded with davdroid, in fact in general having a ROM that came preloaded with a couple of things I actually use for once was refreshing.

The makeovers they’ve given some apps like K9 Mail honestly feel a long time overdue.

Setting up an /e/ account was pretty smooth and painless and doesn’t lock you out of adding subsequent email / caledav / carddav/ webdav accounts either.

You can ignore the vast majority of the /e/ bits if you want and just use it as a LineageOS + MicroG ROM.

The Bliss launcher from /e/

Conclusions

It’s honestly too soon to judge /e/ one way or the other. However even at this early stages it begs a few questions. Is substituting the e.fundation for google a good thing?

Is what their adding ontop of LineageOS & MicroG going to add enough extra value?

I am honestly not sure. I’ve gone from optimistic to cautious-but-not-yet-critical as ive tried out the early build. It’s worth continuing to follow but im not entirely sure they have found their feet or entirely the best direction to set off in yet.

Addendum: I switched back to vanilla LineageOS + MicroG a few days after writing the bulk of this. Bliss has some bugs in terms of slower microSD cards, with apps installed on the external storage that grated.

Not a terrible sin for such an early build, but such things may make it unsuitable as one’s daily driver just yet.