Month: December 2017

Home with the Sniffles

I caught some kind of head cold, or maybe the flu that's going around, and stayed home from work today.

Reports indicate that this year's flu vaccine may only have a 10% efficacy. But if you haven't been vaccinated, do it; 10% efficacy's better than nothing. I got mine -- and if I hadn't, I might be feeling even worse right now.

Android Without Google

In my last couple of posts, I've talked a bit about the drawbacks of iOS and Android, but acknowledged I've found the alternatives lacking. Ultimately, I went back to Android -- but not stock Android.

Android -- at least, the base OS -- is free/open-source software. As such, there are many different variations of Android available.

Replicant is the only Android variation endorsed by the GNU Project; it seeks to provide an Android experience with only free/open software. Unfortunately, it has drawbacks: it has a very limited number of supported devices, the most recent of which is the Samsung Galaxy Tab 3, which was released in 2013. Replicant itself isn't quite that outdated; the latest version is 6.0, based on Android 6.0 Marshmallow (2015). And even though Replicant itself is free, it still requires proprietary firmware in most cases.

I've ultimately settled on LineageOS, an Android distribution descended from the previous CyanogenMod project.

You can install Google Services and Apps (Gapps) on top of LineageOS, but on my latest installation, I opted not to do that. I get most of my Android software from F-Droid, a free software repository.

I do run a few proprietary apps; the Amazon App Store is one source, and there's a program called Yalp Store (you can get it from F-Droid) that lets you download apps from the Google Play Store without installing Gapps -- though keep in mind that does violate Google's terms of service.

Someone also recently recommended microG to me; it's a free re-implementation of Google Services. I haven't tried it out yet, but it looks promising.

All in all, I was surprised by just how easy it ended up being running an Android-based OS without Google's proprietary apps and services. That's easy for values of "easy" that include being comfortable flashing your phone, of course, but so far it's worked out pretty well for me.

I'd sure like to see one of those alternatives get a better foothold, though. More competition is good for everybody, especially if that competition comes from free software.

Android Alternatives

Yesterday I talked a little bit about Ubuntu Touch, a would-be alternative smartphone OS based on GNU/Linux (that is to say, the Linux kernel and GNU userland, as opposed to Android, which is based on the Linux kernel and Google's own userland).

There are other phone OS's out there, too.

Jolla's Sailfish is another GNU/Linux-based OS, based on Nokia's abandoned MeeGo platform. It's the most mature of the lot, but supports a limited number of devices. I haven't tried it because the port for my phone, the Nexus 5, hasn't been updated since 2015. But it appears to have pretty good support for Sony Xperia phones, and it runs Android apps through a compatibility layer, though my understanding is that that compatibility layer is proprietary, drains the battery significantly, and doesn't have full compatibility.

Other than iOS, Android, and, to a lesser extent, Windows Phone, Ubuntu Touch, and Sailfish, there aren't a lot of mobile OS's that are ready for prime-time. KDE's Plasma Mobile is still in early stages; the steps for setting it up on a Nexus 5 indicate that it's strictly for developers right now.

GNOME doesn't have much of a mobile presence at this time, either, though Purism has announced that its upcoming Librem 5 phone will feature a GNOME desktop (with Plasma as an alternative option).

There's also LuneOS, a fork of Palm/HP's webOS (which, like Android, is based on the Linux kernel but not the GNU userland). It's still early days too.

I also just ran across postmarketOS, whose homepage says "The project is at very early stages of development and is not usable for most users yet." (Boldface in original.)

One of the biggest problems facing all these projects is the proliferation of different Android devices, most of which rely on proprietary firmware for hardware support. There is a project in the works that should help with the hardware support issues (though not with the inherent problems of proprietary firmware); it's called Halium, and it should make development much easier for all these projects.

In the meantime, though? You're probably stuck with iOS or Android -- Apple's walled garden or Google's spyware.

There are ways to run Android without Google services or proprietary software. I'll get to that tomorrow.

The Sorry State of Smartphones

It's disappointing that the smartphone market has turned into a choice between two OS's: iOS's walled-garden approach where Apple decides what software you're allowed to run on the phone that you ostensibly own, and Android's spyware panopticon security nightmare.

There are a few alternatives, none of them very good.

A few months ago, I tried switching from Android to Ubuntu Touch. Canonical abandoned Ubuntu Touch a few months back, but it's still under development by a small community-based group called UBports.

Here's what I wrote at the time (originally posted on Brontoforumus, 2017-07-03):

It's a pretty different idiom from Android (no ubiquitous three buttons at the bottom of the screen, though their functionality is there; swipe from the left edge of the screen to get a dock, from the right edge to get a Windows 7-style list of open programs, and the Back button is handled at the app level), but I could get used to it, and the list of available apps seemed sufficient for my day-to-day use.

The only real problem was that the phone didn't work.

I fucked around with the settings for awhile but all I managed to accomplish was to change what it said under "carrier" from "Sprint" to "none".

So I decided to give LineageOS another shot. (Well, technically my first time using it as LineageOS, but I used it plenty when it was Cyanogenmod.) It appears that I've mostly fixed the Sprint issues I had with it before.

But I thought Ubuntu was pretty impressive, and I intend to give it another shot someday. Maybe once they finish updating it to a 16.04 base.

I should probably update my post about getting Sprint to work on LineageOS (then CyanogenMod); I need to update the title and the links, and add the last step that finally got it (mostly) working.

I've managed to do okay without Gapps, too -- but maybe I'll get to that another time.

The Last Day at Papago

Papago Brewing used to be my regular watering hole. It closed on Saturday.

Papago Plaza -- the entire complex where the tap room was located -- is being demolished to put up condos. They've expressed hope for finding a new home, but no news yet.

My good friend Brad -- himself a brewery owner these days -- came in from Riverside to pay his respects, and so we got a Lyft van-full of the old gang together and headed up there for the last day.

There are pictures. I don't have them yet. Hopefully I'll get them later and be able to post them.

It was bittersweet. The fridge was mostly empty; most of the items on the menu, food or drink, were sold out. The life-size monk statue had already gone, as had one of the two dartboards.

The writing was literally on the wall; people had been saying their goodbyes in silver Sharpie for months (if one message, dated March, is anything to go on).

We had a few rounds, and then we walked a block south to McFate -- that's my regular watering hole these days. My friends hadn't been there yet, but they were interested in checking it out. There was a nice bit of symmetry: saying goodbye to the old spot, and hello to the new one.

The details of the day are a little hazy. I remember we told old stories, and I laughed some belly laughs.

I'm pretty sure I only drank five beers, and I paced myself, with a glass of water after each. But I do have a tendency to make a beeline for the highest-alcohol beer on the menu when I don't have to drive. (I can recommend the beers I drank at McFate, but can't remember their names. There was an IPA called Hazy something, and a Scotch-aged something or other.)

I'll miss Papago. I hope it reopens someplace. At any rate, it was good to get the band back together for a day, and talk about the good old days.

Net Neutrality Roundup #2

Yesterday I discussed Ajit Pai's plan to dismantle the FCC's net neutrality regulations, his disingenuous justifications for doing so, and the inevitability of lawsuits challenging the change in court. We left off on Tim Wu's observation that Pai is doing this in opposition to the vast majority of public opinion.

The FCC comment period concerning the net neutrality repeal saw 22 million comments; it received a greater response than any other FCC proposal in history.

While Pai has openly acknowledged that he doesn't care about the quantity of pro-Title II comments, he has also, disingenuously, drawn a false equivalence between the number of pro- comments and the number of anti- comments. Jon Brodkin at Ars Technica notes:

Pai [...] released a "Myth vs. Fact" sheet that claims public comments to the FCC don't show significant support for net neutrality. Pai's office called it a "myth" that commenters "overwhelmingly want the FCC to preserve and protect net neutrality," arguing that fraudulent comments far outnumber legitimate ones.

That's true largely because the FCC imposed no real restrictions on comment uploads and took no steps to remove fraudulent comments from the record. But analyses of comments show that about 98 or 99 percent of "unique" comments oppose the net neutrality repeal.

That last link goes to an article by Jeff Kao at Hackernoon titled More than a Million Pro-Repeal Net Neutrality Comments were Likely Faked. Here are a couple lines from the abstract:

My research found at least 1.3 million fake pro-repeal comments, with suspicions about many more. In fact, the sum of fake pro-repeal comments in the proceeding may number in the millions. In this post, I will point out one particularly egregious spambot submission, make the case that there are likely many more pro-repeal spambots yet to be confirmed, and estimate the public position on net neutrality in the “organic” public submissions.

Kao goes on to chart the duplicate versus unique comments:

Chart of trends in FCC comments

Keep-Net Neutrality comments were much more likely to deviate from the form letter, and dominated in the long tail.

From this chart we can see that the pro-repeal comments (there are approximately 8.6 million of them) are much more likely to be exact duplicates (dark red bars) and are submitted in much larger blocks. If even 25% of these pro-repeal comments are found to have been spam, that would still result in more than 2 million faked pro-repeal comments, each with an email address attached. Further verification should be done on the email addresses used to submit these likely spam comments.

On the other hand, comments in favor of net neutrality were more likely to deviate from a form letter (light green, as opposed to dark green bars) and were much more numerous in the long tail. If the type, means of submission, and ‘spamminess’ of comments from both sides were equal, we would expect a roughly even distribution of light and dark, red and green, throughout the bars. This is evidently not the case here.

Kao has gone to more trouble than Pai to try and tell spam comments from legitimate ones. Indeed, New York AG Eric Schneiderman has accused the FCC of stonewalling his investigation into FCC comments using fraudulent names and addresses. And he's not the only one investigating:

Schneiderman is not the first to accuse the FCC of stonewalling investigations into the net neutrality comment system. The FCC's claim that the comment system was temporarily disrupted by DDoS attacks has received lots of attention, but the FCC hasn't provided all the records requested in several Freedom of Information Act (FoIA) requests.

The FCC also told members of Congress that it won't reveal exactly how it plans to prevent future attacks on the public comment system.

A FoIA request from Ars was denied by the FCC due to "an ongoing investigation."

US Senator Ron Wyden (D-Ore.) criticized the FCC for failing to turn over its internal analysis of the DDoS attacks that hit the FCC's public comment system. Senator Brian Schatz (D-Hawaii) and Rep. Frank Pallone (D-N.J.) requested an independent investigation into the DDoS attacks, and the US Government Accountability Office (GAO) has agreed to investigate.

The FCC is also facing a lawsuit alleging that it ignored a FoIA request for data related to bulk comment uploads, which may contain comments falsely attributed to people without their knowledge.

So are there any other ways the FCC's terrible plan is vulnerable to litigation?

Why yes. Yes there are. Because it also prevents states from passing their own net neutrality laws.

And there's legal precedent stating that the FCC can't preempt state laws -- ironically, decided in a case where Tom Wheeler's FCC attempted to prevent states from passing laws against municipal broadband.

These are just some of the avenues of attack Pai has opened himself up to. The question isn't whether there will be lawsuits after the net neutrality repeal; it's how many and how soon.