I’ve been a very poor care taker of you. I post once a month, maybe; and, when I do, I post uninteresting stuff like the tale of my trip back to California. I don’t, through you, even tell the story of my trip to Berlin, Belfast, or Wales, or even about my trip ideas to Switzerland, France, Singapore, and Thailand. I didn’t even bother to tell you of my trip back to my homeland (to Florida) to see if I might catch a glimpse of the Space Shuttle piercing the sky on its final mission to the stars.

What I have done is ignore you. I don’t even fix my Blogroll links, a substantial portion of which have gone stale in the years since I’ve updated them. Hell, my About page still says I live in Colorado! If it weren’t for the security holes in prior releases, you’d still be running an ancient version of WordPress.

My online presence seems to circle around Twitter and Facebook nowadays, links to which I don’t bother posting in some vain attempt to prevent someone from very easily producing a complete graph of my Internet existence.

What am I going to do with you?

I have no idea. But I think I can safely now consider you “deprecated” until that idea comes.

Until we meet again,

Sean

If you’re unfamiliar with Boxee; a good way to think of it is as a media aggregator. It turns your TV-connected computer into a media center that can access movies and shows from Netflix, Hulu, Fox TV, CBS and music from Pandora, et. al. It’s a cool way to bring your digital media into your living room (and may let you cut the cord from your cable co!)

So, real quick,

The Good

• Neatly wraps up the content sources I like to use: Hulu, CBS, Fox, Netflix, Pandora.
• Works with the Apple Remote
• Provides a way to queue up TV content automagically, to give me content to work through.
• Provides a handy way to provide Boxee-specific proxy settings, for the direct proxy I will use when I get around to setting it up.

The Bad

• Netflix and Pandora are “applications” within Boxee; they live on their own. Separate queues, separate interfaces, etc. It’d be nice to have one queue of TV and movies.
• The TV watching interface is pretty clunky. If it drops out to Fox or to Hulu, it does so inside a browser (that you can see but not drive), and it requires you to press the “any” key to get full screen.
• It does not integrate with my Hulu queue.

The Ugly

• If content is available from 2 sources, say from both Fox and Hulu, it will prefer Fox. Sometimes it gives me the choice between Hulu and Fox, and sometimes not. This is really frustrating, since Fox-over-the-tubes is at a substantially lower bit rate than Hulu (it looks worse).

That all said, Boxee serves about 60% of my needs. When it can’t do what I want, I just drop back out to a web browser and call it good.

If this is a major problem — then I must say, there are some providers on the vanguard of this here.

I just went in for my annual (where annual is defined as 12 +/- 6 months) physical, and my doctor wanted to see a lipid panel (cholesterol screen). I see a doctor at Associates of Family Medicine here in Fort Collins. I casually mentioned that I had a blood draw for just that purpose last week over at Poudre Valley Health System’s Harmony Lab for my cardiologist over at Heart Center of the Rockies. So, my doctor on her laptop, logged into their system and pulled the results — just like that. No sweat at all.

I have to say, that’s not bad for coordinating 2 providers (and accessorily the third provider, my cardiologist).

Still some work to go — a unified patient record would be nice, so I don’t have to mention that I had the test done. But, I have to say, Associates in Family Medicine and the Poudre Valley Health System, at least, are very solidly on the right track.

By the way, my health is good.

You see it all over the place. We choose larger homes, just in case we have kids or get a bigger TV. We buy large cars to cart the kids around or to ensure against the grisly in case of a catastrophic accident. We opt for trucks instead of cars in case we we need to haul a load from Home Depot, or purchase the tow package in case a boat should happen to come into our possession.

We regularly trade cost for the convenience of something we don’t regularly or often have a need for. This is all well and good — we like things bigger, better, and more featureful to make our lives easier.

The thing is, this isn’t true in software development. We regularly plan and develop for the average case — the typical use case — and eschew the exceptional. And I think this is a shame.

Not all organizations do this to the same extreme. However, the desire to get something out to market and the 80/20 rule (or whatever variation you use) often trumps the one feature or slight tweak that could make someone’s day. In software development (and this is probably true of product development in general), we trade box specs for time. We get as much done of a feature as we need to say that we have “it,” then we call “it” done.

This is discouraging.

I think some of this mentality is related to agile development practices. Agile teams, combined with aggressive product owners, move the development team from feature to feature with an ever-increasing myopia towards the requirements of the person yelling the loudest. They’ll tell you that they give as much as the market demands, but in the software world, it’s often (the so-called) business leaders dictating the the “requirements.” I’ve noticed this forgets the people charged with actually using the software, whose own ideas for improvement get drowned out in the “process.”

The end result is a product that can do the typical, a product that can do the average. Certainly not a product that expects the exceptional.

Is this a problem? No, not necessarily.

An average product serves the need of 80% (or more) of the people who use it. That’s a pretty good watermark. What troubles me is how we often overlook the one or two things we could do to make it more like 90 or 95% of the cases.

We don’t plan for the exceptional in software; and I think this manifests itself in a product that doesn’t feel “fully baked” when you use it.

In my rack room, I have a HP ProCurve Gigabit switch powering my network. I love this switch; it’s fast, it has an easy-to-use interface, it has enough ports for all the jacks throughout my house, and most of all: it rarely requires any management time from me. There’s one thing though that irritates me to no end about this switch: you can’t name the ports! This thing has 24 ports on it, but I’m expected to remember that port 13 is my cable modem and port 21 is the file server. To me, adding the ability to name the ports, is low-hanging fruit that would make my life substantially easier. This is the exceptional case; I’m using this switch as the core of my small network, rather than at the edge like HP intended.

Serving an exceptional case doesn’t always have to add a lot of work. In the case of my switch (and I’m going out on a limb here), I can’t imagine adding port naming would’ve been a schedule-blowing feature but it would have made life a lot easier for my case.

I think one of the nice features of Waterfall as a development methodology is that it specifies the whole solution. Agile development is the current industry best practice (for good reason), but it would be nice if we could bring some of the “whole solution” mentality to the process.

I used the ZFS Quick Start Guide to get started. I also cuddled up with “man zpool” and “man zfs” in order to tailor my own solution. In the end, I came up with a relatively simple:

[root@protego]-~# zpool list

NAME   SIZE   USED  AVAIL    CAP  HEALTH  ALTROOT

tank   928G   312G   616G    33%  ONLINE  -

[root@protego]-~# zfs list

NAME              USED  AVAIL  REFER  MOUNTPOINT

tank              312G   601G    19K  /tank

tank/fbsd         497M   601G    18K  /tank/fbsd

tank/fbsd/ports   193M   601G   193M  /usr/ports

tank/fbsd/src     304M   601G   304M  /usr/src

tank/media        203G   601G   203G  /data/media

tank/tm           109G   291G   109G  /data/tm

I did notice that FreeBSD’s default configuration caused kernel panics, because the kernel tries to use more physical memory than exist to store ZFS internals (on a machine with 2GB of RAM anyway). Luckily, you can tune these; I added the following to /boot/loader.conf on the advice of the starter guide to alleviate the problem:

vm.kmem_size_max="1024M"

vm.kmem_size="1024M"

vfs.zfs.arc_max="100M"

So far so good. I’m going to let this run for a while and spend time learning about ZFS snapshots and rollbacks. I’m also going to look into ZFS diagnostics and maintenance; I’m a little concerned about copy-on-write and fragmentation (not such a big deal with the largely write-it-and-forget-it data I have on it now, but something to think about if I say, move /home onto ZFS).

That means it’s time to play around in the technology playground and maybe learn a thing or two in the process. Today’s project: ZFS.

I usually like to pick projects that solve a problem I’m having (even if that problem isn’t particularly grave). In this case, I have about 4TB of storage split across 7 drives of various sizes (and for the moment, split between 2 machines). I’ve arranged the partitions in a way that meets my current needs, but it’s hardly flexible. If I need to reallocate storage, I have to do play musical chairs with my data in order to refactor the layout.

Ideally, I could map all the storage into a large pool and allocate shares dynamically among my various needs: media store, backups, Time Machine, and regular data. I’d like the shares to have different properties, e.g; I don’t care about access times on my media store (noatime). If one share needs more, I’d like to be able to allocate it extra storage from offsets elsewhere. I’m hoping ZFS gets me at least part of the way there.

To start with, I’m playing musical chairs with about 200GB of data. I’m copying 200GB off two 500GB drives (arranged in a mirror) over the wire to my file server (boy, am I glad NFS is easy to setup). Once I’m done, I’ll have 1 TB of free storage (2x500GB) to setup in a testbed ZFS pool (just as soon as I extract the drives from avifors and re-mount them in protego!)

I love geek days.

1 – except my run, of course…

It all started with episkey, my main hosting server. For about 2 years, it had been colocated at Red Rocks Data Center (these guys are great, I totally recommend ‘em), just outside of Denver. After my largest tenant (an IRC server and most of the accessory services for the network) departed, I felt that the cost of the colocation far outweighed the tangible benefit to me; so, I pulled the server back home to live on my network and on my /29 from Comcast.

At the data center, episkey spent its computing hours cooped up inside half-depth 1U case; a case that would have been an ideal fit in my rack at home… if it weren’t for the 10,000 decibels of industrial-grade cooling system blowing out the back. Honestly, if the case had an airfoil and a runway, it’d probably beat “my” 172 to pattern altitude.

So, I ordered a new case (an Antec Sonata III), new quiet heat sink and fan, and a new heat sink mounting bracket (for former enclosure necessitated custom set). I built that all up in the new case and all was well, until: POP (and quite possibly, a few snaps). The power supply emitted (emut? new word, perhaps?) its secret smoke and that, my friends, was all she wrote.

The 500W power supply in the Sonata III was done; totally cooked. Oh, and it took the motherboard with it. episkey, if it were constituted solely by its motherboard, processor, and memory, was now dead. A faint miasma of death (which is astonishingly similar to capacitor smoke) still lingers to this day.

Thus, began the replacement process.

I decided to go “newish” and “low cost” — one thing I’ve realized is that for my needs, I don’t need the fastest out there. This “new” box is going to run FreeBSD.  It’s going to have a steady and predictable workload very similar to what the former Pentium 4 3.0GHz incarnation was able to do without so much as breaking an electronic sweat. So, in a nutshell, I didn’t need the latest nor the greatest. High quality but middle of the road (perhaps a little older, then), would do.

But first, a new case. If you weren’t already aware, Amazon has fantastic customer service. I explained what happened to them and they gladly sent me out a replacement case and power supply. It’s really nice to not have to worry about fighting a retailer for things like this, so kudos to you Amazon. I figured I was on my own for the board, so I started shopping around.

Ultimately, I opted for a used Supermicro board (a C2SBA+II) from eBay — it’s a LGA 775 board, has 6 SATA 3Gb/s ports, and can carry 8GB of RAM. I picked up this “baby” for a song (and $45) on eBay, then added on a 2.6GHz Pentium Dual Core and 2GB of RAM. I also added on a PCI 10base T (yes kids, that’s a whole 10 megabits per second!) card that I borrowed from Myles, as I discovered that my awesome eBay find has a bad Gigabit NIC onboard (PCIe replacement already enroute). All built out, I christened it “protego” (in keeping with my Harry Potter naming scheme) and installed (then -STABLE’ized) FreeBSD:

FreeBSD protego.dreamfire.net 8.0-STABLE FreeBSD 8.0-STABLE #0: Sun May  2 19:06:48 MDT 2010

root@protego.dreamfire.net:/usr/obj/usr/src/sys/PROTEGO  amd64

I like to beat up on machines a bit; and I like FreeBSD’s make buildworld for just that. So, for this evening, protego will be “building the world” in a continuous loop. As a bit of a benchmark, its first world took just a bit over 38 minutes (I believe episkey was in the few hour mark, avifors — my backup server — takes many many hours).

Because I’m ever the tinkerer, I spent a little time looking at the stenciling on the board itself for interesting clues and for things to turn on and off. I discovered the FSB frequency jumpers and a handy table with the settings to force it to 133MHz (533MHz FSB), 200MHz (800MHz FSB), 266MHz (1066MHz FSB),  333MHz (1333MHz FSB), and the ever interesting default setting of “Auto.” It had been set to the default — Auto — which for this processor was 800MHz.

I decided to bump and lock the FSB at 1066MHz. That jump kicked the CPU from 2.60GHz to 3.5GHz. Unfortunately, a make buildworld immediately caused a kernel panic. I went into the BIOS and tinkered with some of the settings; a little extra voltage here, a mysterious 15% addition (seriously, you can overclock it in 5% increments from 0 to 15%; I have no idea what that means either!) there, and presto: a make buildworld carried on… for a few extra minutes, until it panicked yet again.

Oh well. Can’t have it all. At least, it’s working (back in “Auto” mode).

Take this quote (of a quote) for example:

According to several people present, Andy Lees, a Microsoft senior vice president who oversees development of the mobile-phone software business, and his boss, Robbie Bach, explained that Microsoft workers often use rival products to better understand the competition.

I nearly choked on a carrot reading that one. I mean, come on, “to better understand the competition”? I guess it never occurred to them that the iPhone, despite its many flaws, is just a better device?

One of the things I think Apple does very well is to understand its customer, and produce a device to meet most of the desire and do so in a stunning package. Take the original iPhone for example: it only supported EDGE, didn’t have multi-tasking, and if you wanted to copy-and-paste you were out of luck. On the other hand, it had a real browser and a stunning user experience. The iPhone put a computer (or something close enough) into the hands of its customers, and they were willing to overlook the shortcomings. Even today, the latest iPhone 3GS still doesn’t have multi-tasking, an open-market for applications, or tethering (if you’re in the United States) — but people are still buying them like crazy. Even Microsoft employees.

Microsoft has the luxury, because of its position, to be a follower instead of a leader. It’s always worked for them. I think, though, if they’re wondering why iPhones are popular among their employees, they should look in the mirror and realize that the iPhone is just a better device. It’s not about what people say they want, it’s about what people don’t know they want.

One of the major challenges I have is what to do with the mac mini and MacBook Pro. I do a terrible job synchronizing the documents and work-in-process across machines despite trying a few different automated solutions. Almost invariably, the things I work on (read: code) are stored on a machine that I ssh into.

So, I’m trying out something new: I’m going to stop using the mac mini for anything workstation-related. I have an ongoing need for Windows and for a development FreeBSD server (one that I can experiment with and break a lot, not something I’d like to do with my regular FreeBSD server at home). So, the mini now runs headless with a few virtual machines on it and all my work is conveniently centralized on my laptop.

I’m curious to see how this will work out. Hopefully well. This post (besides being, I hope, at least marginally interesting) serves as sort of a mile-marker for this trial.

If you’re looking to roll your own Time Machine server, and don’t want to shell out for one of Apple’s Time Capsules (say, if you want to integrate Time Machine into your existing storage array), this is how you do it:

What you need:

1. A FreeBSD or Linux server with some spare disk space.
2. A network of some sort.

The how:

Note, the installation and activation (1 and 2) steps are for FreeBSD. If you’re using Linux, your vendor/distribution most likely has a package pre-made for you. Try “apt-get install netatalk”, “yum install netatalk”, or consult your distribution’s documentation. Alternatively, you can install directly from source!

1. Install netatalk

   # cd /usr/ports/net/netatalk
   # make all install
   

2. Activate it on boot, by adding these lines to /etc/rc.conf

   netatalk_enable="YES"
   afpd_enable="YES"
   cnid_metad_enable="YES"
   

3. Configure a Time Machine volume, by editing /usr/local/etc/AppleVolumes.default:

   # first is the path, second is the display name, remaining are the options
   /data/tm "Time Machine" options:nostat,tm cnidscheme:dbd
   

4. Start it up:

   # /usr/local/etc/rc.d/netatalk start
   

   On Linux, try:

   # /etc/init.d/netatalk start
   

5. You’re set. On your Mac, just connect to your server (Finder, Go -> Connect to Server, then enter afp://yourserverip), log in, and mount the Time Machine volume. Then just hit the Time Machine prefpane (Apple -> System Preferences -> Time Machine) and turn it on!

For auto-discovery of your file server, you might install avahi and add an _afpovertcp._tcp service, see this link for more info.

Caveats:

Time Machine will consume all the space on the partition /data/tm lives on. It’s a good idea to create a storage-capped volume specifically for Time Machine, so you don’t accidentally fill up your disk with constant backups!