An Apple TV with a big screen? Nonsense, I hope

I’m so tired of the arguments from pundits saying that Apple needs to make a television.

There is little to no profit in television sets; they’re commodities. Let’s take some of the Japanese companies as examples. Sharp, Panasonic and Sony lost a combined $20 billion dollars in 2012. If Sony hadn’t sold off a chemical unit and didn’t have the PlayStation and content, it might be dead. Sharp is unlikely to survive much longer without a merger/acquisition. Panasonic… dunno.

I won’t go into the details, but I’ll be the first to admit that these 3 companies have made many missteps in the last several years that made them bleed a lot more than they should have. Panasonic made some major losing bets like building new plasma panel factories starting in 2003 and dumping a lot of cash into their purchase of Sanyo for what are now struggling businesses (solar panels and batteries). Sharp completely missed out on smartphones (their phone sales have been halved from 2007 to 2012) and outsourcing, and like Panasonic made missteps in solar. Sony’s television business has been losing money for 8 years running.

The problems here are many. For starters, the competition from China and Korea is difficult to overcome. Large LCD panels (regardless of backlighting technology) are now a commodity. LG and Samsung will continue to make TVs with near-slave labor. And despite what some would like you to believe, people are not starting to change their TV sets every 2 to 3 years. Or even every 4 years. The reality is that the recent trend has been to replace aging non-HD sets with HD sets. I hate to break it to the idiots masquerading as experts, but HD has been with us for more than 10 years. It has taken that long for the average consumer to adopt a single HD set. Once the swing is over, the buying will stop again. Sure, they’ll eventually replace each set in the home. But they’ll then stop for a long time, unless the sets fail. And it would never have happened in the first place if the prices hadn’t dropped dramatically in the last 5 years.

Look at the trend for PCs. It’s not significantly different than televisions. Apple is a publically traded company and they’re making their money in mobile (phones and tablets, plus their content stores). The money they’re making from iMac, Mac Mini and the nearly-dead Mac Pro is a very small piece of their profits today.

The fools who equate the HD screen buying to people buying smartphones every 2 to 3 years should be fired from their writing positions. Tip for the ignoranti: phones are an order of magnitude less costly than 55″ HD televisions. They’re also with us 24 hours a day, and your average smartphone is an order of magnitude more useful than a 55″ television simply due to the fact that it fits in your pocket. They’re also an order of magnitude easier to recycle. But more importantly… most people aren’t buying phones every 2 years just to get a new display. They’re buying a new phone to get new features, often only software features or software features supported by a small bit of updated hardware (say LTE or accelerometer or GPS). And of course to replace the phone they’ve beat up by dropping it, spilling their McDonalds on it, and using it. And to get a new battery.

I should also point out that for many of us, the television is no longer our main consumption device. We’re consuming much or most of our content on tablets and our phones. In part, this is just due to convenience. It’s also in no small part due to being busier than ever. And nearly all of us are using our televisions as simple HDMI targets (often only HDMI video with the audio going to a home theater receiver). What makes the big difference in our experience is the HDMI source, not the HDMI target. Most, if not all of us don’t want to replace a $1000 to $3000 device every few years just to get a decent bump in the source and maybe a small bump in the display. The people who think we’d like to do this are probably the same people that thought refrigerators with built-in smart displays would be a game-changer in the kitchen. Why would I want a smart refrigerator if I can’t put it in my pocket to take it with me to the grocery store? How was that ever a better idea than putting Grocery iQ and Red Laser on my iPhone?

And of course Apple understands these issues. It’s part of the reason that the MacBook Pro has no ethernet port and no built-in CD/DVD player, but instead has USB 3.0 ports and a Thunderbolt port. All the smarts inside, but none of the non-essential parts that would take up space and weight. Of course, I consider an ethernet port essential, but not a built-in CD/DVD player. It’s part of the reason Apple has AirPlay… I don’t need or even want my TV to manage content; I want the device that’s in my hands or pocket or bag all day to be the device that manages my content. Even better if it can hold a lot of it too. I don’t care to have to use my TV to manage content. I want to be able to do it anywhere, anytime.

The difficult battle for a game-changing event in the TV world isn’t over the TV. It’s over the content and the delivery and the contracts that bind the two together. The same reason that the game-changing event in the music world wasn’t the iPod or any other mp3 player, it was content in the iTunes store and other stores like Amazon.

The cable companies are not going to hand over their bandwidth to any company that directly competes with their offerings. And most of the content delivery companies, who also own nearly all of the last mile and then some for residential Internet service, move like snails. And make much of their money with bundled content packages. Anything that disrupts their cash flow is going to be met with loads of resistance. They’re already using caps in many places, and if it weren’t for the net neutrality laws, they’d be blocking all streaming and downloads that they viewed as competing with their offerings.
In fact they’ve done that before (Comcast P2P fiasco, for example) and would still be doing it if there had not been an uproar with some legal footing.

I don’t need or want a TV with a lot of built-in obsolescence. Maybe some people do, but those people represent a tiny piece of the market (those with tons of disposable income for home electronics). What I want is a better content management device and a screen that’s about as dumb as possible that will last 10 years. Keep the screen for a long time, and update the content management device frequently at a much lower cost. The existing Apple TV is a pretty good device at $99. Replace it with something about halfway to a Mac Mini, and you’d have a device more than capable of receiving, storing, managing and delivering content in just about any form. But it’s a problem that doesn’t exist until that content is readily available in a form that such a device could use. In real time as it’s broadcasted, not a week later. And Apple isn’t currently in a position to fix the entire problem. Even if they spent tons of money to acquire the rights to the content, they won’t be able to deliver it if it competes with the last mile provider’s services. And there isn’t a snowball’s chance in hell that they themselves can deliver a significant live event, say the Superbowl. Even if they could, it would never fly past the last-mile providers who’ve paid big money to be the providers of such content.

I’m all for a better home entertainment experience. But an Apple TV with a screen, in and of itself, isn’t going to change the game. And it’s unlikely to make money. It also won’t be significantly better than sets from the established television set producers. 4K display? Who cares at this point given that there’s essentially no 4K content to consume and 4K sets start above $10,000 at this point? 4K will get here eventually, but its audience will be limited for a long time. Most of us don’t need an 84″ or 110″ screen, and many of us would never get it past a significant other even if it was cheap. Well, there’s an argument in Apple’s favor… if you’re going to have a 110″ screen, it’d be easier to get spousal approval if it looked like a piece of art. The problem is that it’ll cost more than $20,000 at this point, and it’ll likely depreciate to $1000 in 5 to 10 years.

In my opinion, anyone that argues that we’ll all be replacing our big screens in our family rooms every few years is a shill or just clueless.

If Apple really wants to change the TV game within the next 5 to 10 years… make a set-top box that has all the capabilities of the Apple TV and more, plus the ability to replace the DVR or cable box that we’re using now (including handling encrypted channels, not just clear QCAM). I know they’ve explored this arena, but haven’t broken the barriers yet. Maybe they will, maybe they won’t. I think our best hope is that the cable companies eventually provide all content over IP (presumably IPv6) and back out of the set-top box business. Don’t hold your breath; those rentals are a source of income for them. One could argue that they could just increase the price of their service and ditch the set top box rentals, but… today they do both (price increases on their service and continued reliance on set-top box rentals).

Now running Clover bootloader on my hackintosh

I’ve been wanting to try the Clover bootloader on my hackintosh for quite some time. I just didn’t have an impetus to bother with it. That is, until the issue with iMessage appeared in late December 2012 and the various Chameleon developers and their branches did nothing about it (including Chimera, which is what I was originally using before switching to Chameleon and now Clover).

Clover is working great for me, and I don’t see myself going back to Chameleon or a Chameleon derivative like Chimera. The reality is that development is MUCH more active on Clover, and given that it has much better support for UEFI (including a built-in EFI shell), it’s the right thing for me to use on future builds. Plus it already has the infrastructure to cope with issues like the iMessage snafu.

Since I consider this switch permanent, I spent a little time creating a Clover theme since I wanted an orange-colored theme. Here’s a screenshot from the theme package I uploaded to insanelymac.com:

Soapbox on…

I feel sorry for the many people on the tonymacx86 forum that don’t seem to realize that moving from Chimera to Clover is not terribly difficult. The bootloader is a tiny portion of what they’ve gotten from tonymacx86 and Chimera is not significantly different than a vanilla Chameleon build. In fact I had been using Chameleon for the last few months before I switched to Clover. The last time I diffed Chimera and Chameleon source tree trunks, the differences were purely cosmetic. I feel really sorry for the folks who gave money to tonymacx86 recently for the fixes for iMessage, given that the tonymacx86 team did NOTHING to resolve the issue. AS of today, that thread on tonymacx86 is 68 pages and growing, with almost 100,000 views. The fix is to use Clover, and the tips on getting that done were all provided by members, not the tonymacx86 team. I’m starting to understand why some other sites’ members despise the tonymacx86 site. You can’t blame the members of the tonymacx86 site, they’re completely innocent. And I guess I can’t blame the tonymacx86 team for the ignorance of some of the members that have given them money, but… it still feels underhanded to me. Especially given the fact that the tonymacx86 tools are primitive and somewhat braindead. MultiBeast might work for some users who really don’t know what they’re doing (and I was once one of those people), but at the same time, stomping on kernel extensions without showing the user what’s being done (and not providing a means for them to revert the changes) is downright user-unfriendly and amateurish from a software engineering perspective.

Site Health page now shows UPS status and history

I am now collecting UPS status data from my UPS, and there is a new plot on the Site Health page that displays it. I still need to make these plots work more like those on the Site Traffic page, but having the UPS data for battery charge level, UPS load, expected runtime on battery and power consumption is useful to me. I currently have 3 computers plus some other gear running from one UPS, but soon will move a few things to a second UPS to increase my expected on-battery runtime a bit.

new computer rack: Startech RK2536BKF

I recently decided it was time to retire one of my 42U computer racks. It was overkill for my needs, and I didn’t look forward to having to move it in the spring.

I replaced it with a refurbished Startech RK2536BKF, which is a 25U rack with a vented glass front door and a mesh rear door. It is near perfect for my needs, and I couldn’t pass up the price: $386.99 plus $111.55 shipping from Startech via Amazon. I could not identify anything on the rack that would indicate it is refurbished; it looked new to me.

Here’s a picture of the front of the new rack with the door open. From top to bottom: ART PS4x4 power distribution, Middle Atlantic TD-5 5U drawer (for rack screws, ethernet cables, etc.), an HP TFT76000 KVM console (17″ LCD, keyboard and trackpad), a Belkin OmniView PRO 8-port KVM, ria and www on a shelf, depot, a Powerware 5125 1500VA UPS and a Powerware 5115 1400VA UPS. Not visible are the parts on the rear of the rack: a Zyxel GS1100-16 gigabit ethernet switch, another ART PS4x4 PDU and a pair of Middle Atlantic QTFP-2 fan panels.

I will be putting my 42U rack on craigslist. It’s also a Startech breakdown rack, with an acrylic front door, steel rear door and 3 fans in the top.

depot: new backup/media server

I’ve slowly but surely been working on a new server in my rack. The intent of this server is to consolidate copies of some of my backups, and provide a place to store media files (music, movies, etc.). This new machine is known as depot.

The intent is to run FreeBSD 9.1 and ZFS. I will likely start with a single pool of 6 drives in raidz2, and later add a second pool.

As part of this process, I’ve migrated to a Startech RK2536BKF rack.

Microsoft’s Windows business model still unfriendly

A story I’m living at the moment…

My uncle’s computer died. Through my mom, I wound up with the task of fixing it, moslty in the interest of trying to recover whatever I can from the hard drive.

The machine is an HP Pavilion a1010n. It did not ship with recovery CDs. Instead, it had a recovery partition on the hard drive. Not useful when the hard drive is in very bad shape (all partitions have unreadable blocks, including of course the recovery partition).

I checked HP’s web site. Recovery CDs are no longer available, thanks to Microsoft’s draconian licensing to OEMs. I installed Windows XP on a new hard drive using a torrent download of Windows XP Home, just so I could boot it and try to run the recovery CD creation tool on the original hard drive. That didn’t work due to too much damage on the original hard drive; creation of the recovery CDs failed on the first CD. I tried activating the Windows XP on the new hard drive using the machine’s original XP activation key. That also failed.

Microsoft doesn’t let you download Windows XP. Doesn’t matter how many times you’ve paid for it. You can’t get a copy from Microsoft at all, by download or CD or DVD or USB stick or any other form.

I’m in the same boat… I paid for a copy of Windows XP Pro years ago, but I can’t find the CD. I have the flimsy tri-fold box and the activation key, but the CD fell out of the flimsy holder and I can’t find it. So the activation key is completely useless and there’s nothing I can do.

This is only the tip of the iceberg. I’m not a novice computer user. I’m a software developer with over 20 years of experience. I wisely abandoned Windows a long time ago, but for a very few things that only run on Windows that I need (BMW TIS, BMW ETK… that’s probably it). Today, I never install Windows natively; I only run it in a virtual machine under vmware, and very rarely. Why? It’s ludicrously expensive, it’s confusing for the average person to buy (am I eligible for an upgrade version? Do I have to spend big $$$ for a full version? Will a home version suffice or do I need a ‘pro’ or ‘premium’ or ‘ultimate’ edition? OEM or retail version? How does the average consumer know?). Worse, in the end it’s an inferior operating system to freenix (FreeBSD or Linux) and vastly inferior to OS X. FreeBSD and Linux are free. Of course I’m only comparing operating systems here, not applications. We all need the operating system(s) we use for the set of applications we need to run on top of them, and usually make our operating system choice based on the applications we need and the applications we already own (though often we wind up having to buy those applications again if we’re upgrading a commercial operating system).

Let’s compare to Apple. OS X Mountain Lion? $19.99, for a full version, downloadable, and available in the cloud once purchased. Download it as many times as you need. Save it to a USB stick, hard drive, backup, DVDs, whatever. Install it on all your personal machines. Again… $19.99. Instant gratification, fast and easy installation. No media to lose or damage. No activation key to lose. No obnoxious pricing-based-on-removal-of-features. No hassle when you need to install a new hard drive and reinstall your operating system and all of the applications you bought through the app store.

If Microsoft wonders why Apple is so successful these days, they should buy some Apple machines and use them for a few years. Perhaps they’d realize that when you make the consumer’s computing life easy instead of difficult, you wind up with customers who never want to leave. Instead of customers who wind up feeling “There MUST be something less painful for me to maintain”.

I completely abandoned Windows around the start of Windows XP and never looked back. Where did I go? Initially, I went to UNIX in various forms. I continue to run freenix machines for all of my server needs. I didn’t come to Apple until OS X, and in some ways it was a roundabout… I loved NeXT in the late 1980’s and early 1990’s. I was never a huge fan of MacOS; it was OK, but didn’t fit my needs. When Steve Jobs came back to Apple and brought NeXTSTEP with him, I knew I’d have the best of both worlds in OS X. Great user interface on top of a modern operating system that’s far more mature under the hood than Windows. With free development tools that are nicer than anything from Microsoft. While Apple has historically probably had as rocky a relationship with developers as Microsoft, that’s not true today and hasn’t been for a long time. OS X is vastly more developer-friendly than Windows. Xcode full-blown version? Free. Interface builder (for both OS X and iOS)? Free, and head and shoulders above Microsoft offerings. Toolchain and standardized language support (say C++)? No comparison.

C# is a nifty language. But it’s not significantly niftier than java, and while you can use it on other platforms, it’s of no real benefit on other platforms; Microsoft’s GUI frameworks are just as closed as Apple’s. CLR? Please. Every product whose guts I’ve seen using a combination of C#, C++ and perhaps some Visual Basic is a mess under the hood. IF you ask me, no one should promote the development of software in three or more languages that’s glued together by a runtime whose idioms and quirks are exposed to the programmer and require non-standard modifications to a standardized language. It’s not that these technologies can’t be used effectively and cleanly. The issue is that in the real world, inexperienced developers are going to use them the wrong way, and as crutches to “make it work” using the tools they know how to use. So you wind up with Visual Basic programmers staying in Visual Basic, C# programmers staying in C# and C++ programmers staying in C++ (and bemoaning the nightmares of the work required to either isolate standard C++ code or convert it to managed code). The glue is usually very messy and riddled with problems that put dire constraints on refactoring because experts in all three languages with a keen eye for proper interfaces and integration are few and far between.

It’ll never happen, but if Apple decided to sell OS X for non-Apple hardware, Microsoft would likely be in big trouble in under 5 years. Just look at the explosion of hackintosh usage over the last few years. Look at tonymacx86.com for an example of how many users (including complete novices) have jumped on the hackintosh bandwagon.

Microsoft could fix most of these issues fairly quickly. Will they? I don’t know. Windows 8 still looks pricey to me, but at least it’s less absurd than previous releases. And I like that they’ve done a ground-up UI redesign, and applaud the courage of doing so. But if they continue to trouble users on repairs, upgrades, edition confusion, etc., they will never win over an experienced OS X user. In this day and age, I should never need physical media nor an activation key to install, upgrade or repair a PC. The issue for Microsoft is that the flow of users between Windows and OS X is almost completely unidirectional… users go from Windows to OS X and almost never go the other direction. And it’s getting worse because Apple has had recovery and syncing mechanisms for all of their products for a long time now, that don’t require a bunch of effort and discipline from the user. This matters because today, we’re keeping an awful lot of the history of our lives as bits instead of ink. All of my contacts are in my iPhone (and backed up in the cloud, on my PC, on my laptop and on backups). Most of my photos are digital and have been for a decade. Receipts? Most of ’em are in an email folder (though I have paper copies too). So little things like Time Machine and iPhone syncing wind up being critically important to the long-term user experience.

Do not get me wrong… I want Microsoft to fix these issues and to succeed. I want competition in the operating system market. Not just in dollars and popularity (Windows’ market share continues to be much larger than OS X’s), but in power, ease of use, ease of maintenance… consumer experience. Today, OS X provides the best long-term user experience, by a wide margin. I want choice for those of us who want and need a high-quality user experience. As a developer, I want operating systems that are a pleasure to use and a joy on which to develop software. Let’s hope that Windows 8 and onward start moving in that direction at a better clip than what we’ve seen from Windows 95/98 to XP, XP to Vista and Vista to Windows 7.

New sunglasses ordered: Serengeti Trapani

I admit to being somewhat of a sunglasses snob. It’s not the fashion that I care about… it’s all about the lenses.

I have almost always had a pair of Serengeti’s in my collection. Why? If you ask me, they make the highest quality lenses available for driving, boating and flying. My one complaint about my most recent pair… I went with the Cielo with PhD lenses, frameless, for incredibly lightweight sunglasses. They’re fantastic to wear, but eventually crack around the nosepiece because without frames, the lenses themselves provide all of the structure.

I was down to just my cracked Serengeti Cielo and a pair of Ray Bans. Both broke recently, and now I have only a cheap pair of Oakleys that I don’t like. So today I ordered a new pair of Serengeti Trapani with the polarized drivers lenses. I look forward to receiving them. The lenses are glass and the frames are a titanium and acetate mix; strong and durable, but with some give.

In my experience, a good pair of sunglasses will typically last 5 years or more. That’s with daily use (I never drive in daylight without them). So while $200 seems like a large sum of money for non-prescription sunglasses, I value my eyes. $40/year for something I wear every day is very reasonable. And the key to the longevity… I use straps so I never drop my sunglasses. Fashion faux pas, but I like having quality sunglasses and making them last as long as possible. It’s also the environmentally friendly thing to do versus tossing a cheap pair of sunglasses in the trash once a year.

I like to have more than one pair of quality sunglasses. I find two pairs a necessity and three pairs is better. My eyes are very light sensitive, and hence I always need to wear sunglasses when outdoors; they’re not optional for me. And different lenses serve different purposes. For example, I like the Serengeti Driver’s Polarized lens (glass or PhD, both are exceptional) for summer driving. But I like the CPG lens in harsher light conditions, for example on or near the water and when there’s snow on the ground. That means I prefer the CPG lens during the winter months in Michigan. I like the sportier PhD-lensed models for active wear, but prefer the rimmed glass lens models when socializing or wearing to the office where I’m taking them off frequently and possibly not being very careful with them. For long road trips, I prefer the PhD lenses; the light weight means I can wear them for many hours on end without leaving painful imprints on the bridge of my nose. For longevity I’ve had better luck with the models that have framed glass lenses.

I put two pairs of Maestrale on my birthday wish list, and they are partially-framed PhD lenses. If someone buys them for me, they could easily turn out to be my favorites. They have the benefit of the light weight of the PhD lenses and not a lot of frame, but probably enough frame to prevent me from cracking the lenses at the bridge for a good while. They’re also a larger fit than the Cielo were, which should help with respect to stress on the lenses. I put both the CPG lens and the Driver’s lens on my wish list. I’d love to have both, and will buy them myself if no one buys them for my birthday.

The Trapani are likely too heavy for me to wear for an entire day due to having glass lenses. Hence they’re unlikely to be worn when I’m doing a full day of yard work, on a road trip, or on weekends. In all likelihood they’ll mostly just be used for my work commute. I might just leave them in my car most of the time, which is what I used to do with my metal-framed, glass-lensed Ray-Bans to be sure I always had a pair of sunglasses in the car. The Trapani don’t have the peripheral coverage of the Maestrale, which means they don’t offer as much protection. Even though they cost more than the Maestrale, I’m going to call them my beaters since they’re probably a little more durable.

Secured my mail server

I’m near done configuring my mail server. Last night I got sendmail configured to use STARTTLS, and to require it for SMTP AUTH. I now don’t allow cleartext passwords, so I can feel safe using my iPhone to send mail through my server when I’m not at home. Not enforcing STARTTLS wasn’t a big deal for my desktop since it’s on a secure wired LAN with my mail server, but there are times when I want to use my iPhone and laptop to send mail when I’m away from home, and hence I need to enforce crypto for SMTP AUTH.

All works fine using Mail on my hackintosh, Mail on my MacBook Pro, Outlook on my hackintosh, and of course my iPhone. I need to write up everything I did so I can repeat it in the future if necessary.

running my own mail server again

Many years ago, I ran my own mail server for my domains. I eventually stopped, mostly because I wasn’t happy with the amount of effort it required. I also wasn’t happy with the mail user agents that would easily work with it at the time in a secure manner.

Last week I decided I wanted to set up mail service again, mostly because I missed having explicit control and ownership of my mail. So I configured sendmail once again ay my mail transfer agent, and for IMAP I’m using dovecot. I’m using a chain of procmail and dovecot’s deliver as my sendmail local delivery agent, to drop my mail in Maildir format. I’m not switching over to it just yet, but it appears to all be working after many days of configuring and tweaking. I won’t say this is easy; getting it all working after years of not doing any mail administration involved a fair amount of trial and error.

The good news is that I will soon be reachable via a new, shorter email address and will in all likelihood switch to using it as my primary address. Now that I have OS X as my desktop and an iPhone as my smart phone, I have no MUA issues. I don’t need a web browser interface to my mail, though I may install SquirrelMail at some point just to test it.

One of the issues I ran into while doing this… dovecot wanted some newer versions of libraries I already had installed. Those libraries were dependencies of a LOT of software I had already installed from the FreeBSD ports tree. Hence the whole process took me much longer than expected, with a lot of midnight oil burned when portupgrade failed to do what I needed.