Success, EtreCheck Info – But Not Done Yet

[Chaos Manor Reviews returns after another long hiatus with this series from guest columnist Alex Pournelle, who continues his father’s tradition of ‘doing things so you don’t have to’.

In this second installment (first installment here), Alex continues with his story about solving the slowdown on his MacBook Pro. The last installment left us with Alex doing a reboot. Comments are welcomed at the end of this post. – Editor]

Success!

App Store then ran normally, though it took several minutes to open the updates tab—not surprising, since I had deleted its cache and it needed to catch up on what was actually installed. Practice patience; check e-mail; read a newspaper… Voila: MacOS 10.4.4, iTunes and Camera RAW formats all needed updates. I started the 10.4.4 updater, ensured the Mac actually rebooted, and then went off to do something else for half an hour. That’s from experience: More than a few times, I was certain the updater was running, only to find out one last command box needed clicking, before the update would proceed.

While off raking the front lawn, it occurred to me that this whole dance was familiar: It was quite like fixing Windows Update on XP or 7, when no updates are ever found or Update just doesn’t run. True, Windows doesn’t store that sort of system info in the BIOS, and putting Windows Registry entries to rights (Or slaying them wholesale) isn’t quite the same as deleting plist items, but the overall process was quite similar.

Upon my return, the Mac was almost done—at least it hadn’t hung mid-update. (To their credit, Apple is pretty good at recovering from that rare occasion.) Time to make lunch, keep an eye on progress… There it is: The final reboot, followed by—a black screen.

The Reboot to Black Screen

That reminded me why I ritually disconnect the second screen on my Mac during updates. Every time I reboot after an update, the primary screen doesn’t display the list of users, as usual. (You only see this list when your Mac doesn’t auto-login to one user at startup, much like Windows.) The MacBook Pro responds (caps lock light goes on, screen is lit, sound volume chirps when adjusted), but no video. Don’t Panic. Disconnect the external monitor, and the login screen comes up just fine. For me, this only happens after a system update—not just regular restarts, which I see happens to others. It’s odd, and never happens (to me) in normal operation, just after updates.

So, now, is my computer behaving better? Yes. There are fewer odd pauses. One of the more infuriating was opening a Keychain Access dialog box, clicking on the “Show Password” box, and waiting a good two minutes for the password to come up. By stopwatch, 5 seconds now. (The jury’s still out on Microsoft Word, though—more about that when I have more data.)

I still don’t know why I have dozens of AppleSpell processes careering around my computer, irresponsibly sucking up resources and polluting Activity Monitor with dozens of instantiations. (Specifically, I have one or two copies of AppleSpell.service and four to forty of AppleSpell, which appear to be child processes.) To be fair, none appear to take an appreciable amount of CPU (0.0%, usually), and in-line spell-check seems to be working, but this is annoying. (Far more copies appear when running Google Chrome vs. Firefox or Safari, even with the latest versions.)

OS-based spell-checking (AppleSpell) is controlled in Keyboard System Preferences (Mac for “control panel”), text tab. Interestingly, turning off “spell check everywhere”, then closing the Keyboard menu doesn’t kill any of the AppleSpell processes. I’m sure that a reboot would kill them all, but I actually like the automatic spell-check; even though it’s often unhelpful, it makes me think twice whether my colloquial or neologic affectation is appropriate.

My support-site walkabout did stumble across an incredibly useful tool for capturing, in depth, what your Mac is doing. EtreSoft’s freeware EtreCheck tabulates a very complete and anonymized record of what’s started (Kernel Extensions, Launch Daemons, Internet Plug-ins, etc.), attempted to start, taking up memory, etc. It provides great hardware detail: My laptop’s battery, serial number D863155X08ZDGDLB4, has 133 cycles on it, and listed as in “normal” health. It even provides pre-populated, clickable web searches for, e.g., what “com.valvesoftware.steamclean.plist” or “com.distortedvista.istatmenusprocessserver.plist” might be. My primary backup disk was last accessed 15 days ago—time to run another Time Machine backup, once I’m sure everything else is stable.

EtreCheck

EtreCheck is the de facto standard for submitting information on your Mac to support websites, and you can’t beat the price. It gave details about what was starting up (and not starting up!) in my computer, about which more next time—data nowhere near as easy to divine from Activity Monitor or Login Items (System Preferences | Users & Groups | Current User | Login Items).

Of course, EtreCheck isn’t as friendly for knowing “What’s in this thing, anyway?”, an honor which goes to Apple Menu | About This Mac | Overview | System Report. System Report opens the System Information utility, residing in the Apps | Utilities directory. SysInfo also provides many different stats, like System Power settings, that EtreCheck doesn’t. EtreCheck is a powerful tool, now occupying a spot right next to Disk Utility in my Dock.

Wrapping Up

The system is now up-to-date, save for iTunes. I’ve left off updating iTunes to Version 12.2, which supports Apple Music. I don’t use iTunes anyway, and reports of suddenly-trashed music libraries give this little urgency for me.

Computer seems more stable now, with fewer unexplained delays or the telltale “fan on takeoff power” sound showing Chrome is chewing 140% of CPU. (Percentage of a single core, not total overall CPU, so it really can go above 100%.) I’m writing this report on the MBP, and haven’t cursed once, so that’s a good sign. I don’t know if the resume-from-sleep problem is cured, since it hasn’t been put into sleep or hibernation, but, so far, much improved.

But, Viruses?

The Avast virus scan is now into its 36th hour, with 4+ million files scanned; it’s been at “100%” complete since hour 2. Avast shows 2,700+ infections, but they’ve been racked up while scanning mail files, so I’m fairly sure they’re benign (to Mac) PC viruses. That’ll be another update.

Next time: The Virus That Wasn’t, software toys in the attic, somnambulant Macs, and archaeological app removal.

FCC Open Internet Ruling: A First Reaction

By Alex Pournelle

fccThe FCC finally released its “Net Neutrality” rules, a good three weeks after the vote. The ruling is over 300 pages, including commentaries and dissenting opinions (About which more later). There will be longer, more knowledgeable and in-depth commentary on the entire ruling; consider this my first take.

Officially titled “In the Matter of Protecting and Promoting the Open Internet, GN Docket No. 14-28”, the online version is here, download here. Unfortunately, a first read suggests it’s the full employment act for communications lawyers, a great opportunity for lobbying, lawfare and rent-seeking by large corporations, looking to gain unfair advantage—the very groups to be regulated.

In general, the FCC regulates best when it regulates least, and when a thousand ideas can elbow their way into the marketplace, then succeed or fail on their own merits. Let’s illustrate how.

AWS-3 and CMRS: (Mostly) Good Examples of Federal Regulation

The run-up to the FCC’s “Net Neutrality” ruling completely overshadowed Auction 97. Better known as the Advanced Wireless Spectrum 3 (AWS-3) auction, this was the biggest offering of radio spectrum in over a decade, with 31 bidders obtaining 1,611 licenses. AWS-3 was a big deal, and a good example of how government can work well, and not so well.

The Commercial Spectrum Enhancement Act (CSEA) mandated the study and (seven years later) reuse/shared use of various radio bands, particularly for cellular-type services. Marketwatchers thought final proceeds from AWS-3 would be in the low billions, but final receipts hit almost $45 billion, for radio frequencies around 1700 MHz, formerly the sole domain of Federal agencies. Timing was good; wireless data usage was already exploding and projected to rise quickly.

The friction, regulatory burden and overhead of government compliance for AWS-3 has been quite low, by design. And it looks like these licenses will be put into use just as fast as the legal difficulties and spectrum-sharing (Some Feds will continue to be co-users) can be worked out. Cell sites will have more capacity to connect calls, surfers, texters and video uploaders.

Commercial Mobile Radio Service (“CMRS”) is the regulatory classification for mobile telephone services, consolidating PCS, cellular and most of SMR. This light touch also let innovation fly: CMRS licensees can (and have) implemented CDMA, straight GSM, WiMAX and LTE, as technologies improved and the market responded. It would have been very difficult or impossible for the FCC to respond to each signaling standard in depth, but fortunately it didn’t have to.

Not every product succeeded: Qualcomm thought they could broadcast television to handsets as a separate product (MediaFLO), discovered they could not, then sold the spectrum to AT&T, who now uses it as straight cellular spectrum. Qualcomm didn’t give up; it’s pushing LTE-Broadcast to, well, broadcast video to dozens or hundreds of simultaneous viewers, this time within the LTE standard, with help, and with live demos.

Market Forces, Market Innovation

None of these innovations could have happened (or not as quickly) with a much stricter, permissions-based governmental approval cycle, instead of the lassez faire regime for CMRS. Adam Smith’s invisible hand works on the Internet, and it works in RF re-use. (Arguably, it hasn’t worked in terrestrial radio, a discussion for another day.)

LTE-Broadcast did need approval, not by the FCC but the 3GPP. The 3GPP sets standards for LTE communications (currently in Release 13). But the 3GPP isn’t a governmental group; it “unites seven telecommunications standard development organizations”, developing worldwide standards without direct governmental involvement. Approval is less political and certainly more market-savvy than the bad, old, per-market RF technology approvals of the PTT era, or pre-Judge Greene AT&T. The sort-of open-market, engineering-centric approach of the various 3GPP working groups have served the public—both US and global—well.

And that’s the lesson: More freedom, particularly fewer governmental regulations, have allowed a rapid advance in communications standards, capital investment in cellular infrastructure, and the battle between Android and iPhone. This let-the-nerds-loose approach set the stage for such astounding improvements as Artemis’s claimed 35X more efficient pCell cellular demonstration, Alcatel’s lightRadio, and SpiderCloud Wireless, just to name three.

Government Rules, Corporate Shenanigans

On the other side: Government regulation. During AWS-3, bidder DISH Network used tiny subsidiaries to obtain small business discounts for their bids, a clever bit of regulatory jujitsu that did not go unnoticed by their competitors.

That bit of rent-seeking illustrates the bigger problem: Corporations, especially in markets with large sums of money at stake, will use every tool they have to gain unfair advantage. They’d much prefer spending a few million on lobbying to a few billion on competition, which is not good for consumers. The more opportunities in the law, the more they will. There are many in this ruling, especially compared to the truly light regulation under CMRS.

That’s the key issue with the FCC “Open Internet” ruling: If a camel is a horse designed by committee, the FCC Trojan Camelid clearly is nosing open the tent flap. The FCC forbore certain regulations on the Internet, but claims the power to regulate as they see fit under Title II. Many commentators have, incorrectly, said “over 700 rules [under Title II] aren’t going to be applied.” That’s incomplete and inaccurate. The current commissioners cannot bind future ones; what’s to say future commissioners—or the bureaucracy—will stop forebearing?

I’m not the only one to say this—there have been many and many a counter-argument made. FCC Commissioners Ajit Pai and Michael O’Rielly voted against the proposal for good reasons. Pai’s legal objections are summarized here; his policy objections here. Verizon released a “Throwback Thursday” response in Morse code, and another one from a vintage 1930s typewriter, in protest to using old law in this brave new world. There has even been buyer’s remorse (Sort of) from Netflix. Frontier Communications (Who’s buying Verizon’s wireline services in three states) says it’s happy with the reclassification, but that was before the regulations were published. It’s also unsurprising, coming from a company used to (Or maybe counting on) Title II regulation for wireline services. Remember, AT&T was perfectly happy with the regulatory climate before divestiture; it took a big sledgehammer to crack open actual telecom competition.

The Big Show Continues

This is just the first inning. There will be lawsuits, stays, further arguments and court cases. In a future article, I’ll dig deeper into what I see as the fundamental flaws in this ruling (including some I don’t see others discussing). I’ll discuss the “Bright line” rules against blocking, throttling and paid prioritization. I’ll suggest better remedies (Spoiler: Competition) and two Modest Proposals for improving the current Internet. I also welcome your thoughts.

Editor’s Note: Comments are open for this page.