Tuesday, November 7, 2017

OpenBSD on a G4 Cube

I just finished installing OpenBSD on my Power Mac G4 Cube and wanted to write down my steps to success in case anybody else out there ever runs into the same issues I had. For the tl;dr version, skip down to the "NFS Success" section.

First, some back story. I've had my Cube for several years now, though I did not buy it new. When my Cube first arrived, I ordered it a brand new 200GB IDE hard drive. I found a 200GB Maxtor that was not quite as cheap as I'd been hoping since by that time SATA had taken over and IDE drives were getting a bit difficult to find. On a side-note, I discovered that people are making IDE SSDs which is just wild. Must. resist. temptation. to. put. SSD. in. my. Cube...

When I began my Cube experience, Apple's planned obsolescence was well into effect and the latest version of Mac OS X had dropped support several years prior. Happily, the FOSS community continues to support these older machines even now. The first OS I installed on my Cube was Debian Linux. I derive some kind of perverse joy from running current software on a computer that came out in 2000 and whose original vendor has long since discontinued support. It's just cool -- sort of like sticking it to the man. It shows the power of FOSS. Nobody can make a computer obsolete when it's running an open source operating system. It also shows why Intel has such a vested interest in proprietary software. If some start-up were to arrive on the scene with a new, better CPU, there would be nothing to stop the market from shifting away from Intel if the world ran on FOSS. But I digress.

I had actually not been using the Cube for the past few years. It had crashed (though, I now believe this was a freak, one-off occurrence), and it scared me since the Cube was hosting my production Subversion installation. So, I moved Subversion over to a more modern computer and put the Cube on a shelf. But, then, about a month ago, I saw this: <cue the ominous music> Debian GNU/Linux 9 "Stretch" to Drop Support for the PowerPC (PPC) Architecture. Well, I couldn't abide that. My Cube may not have been powered on, but I always knew that, with an internet connection and a couple commands, I could have it back online and updated. So, I began looking into my options.

In the intervening years since I first installed Debian on the Cube, my favorite FOSS operating system has become OpenBSD. I actually ran it on my primary work machine for about 6 months which was a fun challenge. (and the subject of another post some day). Happily, OpenBSD still officially supports PowerPC Macs. Moreover, OpenBSD hasn't bloated up like Linux has in recent years (<cough>... systemd ... <cough>), so it figured to actually run reasonably well on the Cube.

The Task at Hand

If you haven't installed OpenBSD before, it's normally a very straight-forward process. I wouldn't recommend it for novices, but if you understand Un*x-like operating systems, disk partitioning, and you read through the awesome OpenBSD documentation, it's really a cinch. Normally. As I came to discover, my Cube wasn't going to provide me with a typical OpenBSD install experience.

The first sign that there were going to be bumps along the road was when I stuck the new newly-burned install CD into the Cube's CD-ROM drive. Sadly, the Cube's CD-ROM drive has gone up to the big computer part bin in the sky. This is especially sad because the CD-ROM drive was a show-stopper in Steve Jobs's original G4 Cube keynote address. But also sad because OpenBSD doesn't support booting PowerPC Macs from USB thumb drives. According to the documentation, you can boot OpenBSD/macppc in one of the following ways:

  • CD-ROM
  • FFS partition (i.e. an existing OpenBSD installation)
  • HFS partition (i.e. an existing MacOS installation)
  • Remote NFS partition
  • HTTP

That last one is a bit misleading. You can download the install sets from HTTP, but you can't boot from HTTP. CD-ROM was out. FFS was out since you have to have OpenBSD already running. HFS was out since the machine had been running Linux, not MacOS. So that left me with NFS. And, ominously, the documentation gives this notice to those who wish to use NFS:

NOTE: This method of installation is recommended only for those already familiar with using BSD network configuration and management commands. If you aren't, this documentation should help, but it is not intended to be all-encompassing.

Gulp...

Not to be deterred, I proceeded on anyway. There is a scant amount of documentation on NFS booting in the install instructions as well as some more in the diskless(8) man page, but the later is not macppc-specific. I was ultimately successful booting NFS, but not before I tried many, many unsuccessful alternatives.

Fail #1

This actually occurred after several failed NFS attempts, but I'm going to take these out of order, like I'm Quentin Tarantino. At my office, there is a room that is basically a PC graveyard with all sorts of defunct hardware. One of the things in there was an old Dell tower -- old enough to have an IDE hard drive and, more importantly, an IDE CD-ROM drive. My first idea was to try and attach the Dell's CD-ROM drive to the Cube's IDE ribbon cable and try booting from that. After a 24-hour delay for ordering a Torx T10 screwdriver off Amazon, I was in business. Except that it wouldn't boot. I'm still not sure why. I tried switching the drive's jumper switch from cable select to slave and that seemed to get the drive to show up in OpenFirmware, but it wouldn't boot.

Fail #2

My next brilliant idea was to install OpenBSD on the Dell, hook up the Cube's IDE drive to the Dell, format the drive FFS and copy the PowerPC bootloader and install files to the Cube's hard drive and try and boot from the internal hard drive. The instructions mention that the installer creates a 1MB DOS partition for the bootloader when a Mac hard drive is single-booting OpenBSD. So, I went and looked at the source code for the OpenBSD/macppc install script and found the line where the partition is created

echo -n "Creating a 1MB DOS partition and an OpenBSD partition for rest of $_disk..."
dd if=/dev/zero of=/dev/r${_disk}c bs=1m count=1

I ran that command, formatted the partition (newfs -t msdos $wd1i), copied the bootloader over. I then created an OpenBSD disklabel partition, and an FFS partition and copied the kernels and tarball sets across. I then patted myself on my back for being so clever.

Of course, that didn't work either. The Cube was able to load the bootloader, but I couldn't get it to find the bsd.rd kernel on the FFS partition. One of the commenters at misc@openbsd.org thinks this was because the FFS partition was formatted on a Little Endian (i386) machine and the PowerPC is Big Endian. So, yeah...

NFS Fail

So, back to NFS. NFS was actually the first thing I tried since it is an officially supported installation method. I didn't (yet) have a second OpenBSD machine to use, so I installed NFS on my laptop which is running Debian. My mistake is that the Debian implementation of NFS didn't cooperate with OpenFirmware. Might have been user error, but, regardless, setting NFS up in Linux is quite a bit more complex than it is in OpenBSD, especially considering OpenBSD's great instructions.

NFS Success

But that's enough about what didn't work. I finally stumbled onto a winning formula. First of all, booting NFS on OpenBSD/macppc requires three components. (The documentation doesn't make this entirely clear, especially if you read through diskless(8).)

  1. DHCP (/BOOTP)
  2. TFTP
  3. NFS

The DHCP server gives the Cube its IP address, it then loads the bootloader (which is a file called 'ofwboot') via TFTP. And then it points the bootloader to an NFS share which contains the kernel. I knew DHCP and TFTP were working on my Linux laptop since I was able to get into the bootloader. DHCP and TFTP need to be running on the same server, but NFS can be elsewhere. As soon as I moved my NFS server to my OpenBSD Dell, things started working for me. If I was starting from scratch, I would run everything from an OpenBSD server.

Here is my dhcpd.conf: (/etc/dhcpd.conf) I was running Linux isc-dhcp-server, but the configuration is the same for OpenBSD dhcpd:

allow booting;
allow bootp;
authoritative;

host cube {
    next-server x.x.x.x;
    option subnet-mask 255.255.255.0;
    option routers x.x.x.x;
    option root-path "/srv/obsd62";
    fixed-address x.x.x.x;
    hardware ethernet xx:xx:xx:xx:xx:xx;
}

'next-server' is the IP address of the NFS server. 'routers' is the IP of your internet gateway (you'll need this if you want to download anything from the internet during the installation process). 'root-path' is the path to the NFS share (should match your share in your exports file). 'fixed-address' is the IP you want to assign to the computer running the install. And, finally, 'hardware ethernet' is the MAC address of the Mac's ethernet card (which you can look up in OpenFirmware:

0 > dev enet
0 > .properties
[...]
    local-mac-address ......
[...]

You'll then need to set up TFTP (which is honestly just as simple as creating a directory (e.g. /tftpboot) and running "tftpd /tftpboot" as root). Copy the bootloader (ofwboot) into your tftp share and everything else from the macppc distribution folder to your NFS share. (As of version 6.2, all of the files in: http://ftp.openbsd.org/pub/OpenBSD/6.2/macppc)

I should mention before you get started that you should make sure there is only one DHCP server running on your network. In other words, if you network is like almost any network, you should probably disconnect your OpenBSD server and the Mac from the rest of the network until the OpenBSD install kernel is booting so you don't have competing DHCP servers. Once the kernel is booting, you can shut down the DHCP server (rcctl stop dhcpd) on the server and hook the machines back into the network if you need Internet access during the install.

Once you've booted into OpenFirmware (hold down Cmd-Option-O-F after powering on the computer), you just need to type:

0 > boot enet:,ofwboot /bsd.rd

and you'll be booting. Everything else you need is covered in the excellent INSTALL.macppc instructions.

The OpenBSD/macppc Experience

I'm really happy that I've gotten OpenBSD running on my G4 Cube. That said, before you launch into a big project, there are some things you should be aware of. First, not all of the ports that are available for OpenBSD/amd64 or /i386 are available on PowerPC. A couple big ones are missing... Firefox, Thunderbird, GIMP, LibreOffice, Gnumeric. None of those run. Since Chrome's V8 javascript engine has some x86-specific code, Chome (and Chromium) is out as well. As far as I'm aware, there is no webkit or gecko-based web browsers that run at all on OpenBSD/macppc. The best thing I've found so far is Netsurf which has its own rendering engine. Netsurf is a currently-maintained project, but it's a far cry from webkit and gecko-based browsers.

But, if you're like me, this is OK because I plan to primarily run the Cube as a server and all of the server programs I've looked for are there and functional. If you like hacking on things, be sure to join the ports@openbsd.org mailing list and start working on Firefox!

Friday, September 17, 2010

Web Browsers

It has been exactly one year and one day since my last post here. I just read through my previous post and found that I was remarkably prescient concerning the Rangers 2010 success:

Barring a miraculous comeback, this Rangers team appears to be headed to a 9th consecutive season without a playoff appearance. It appears, however, that they have very good chance to stop that streak before it hits 10.

This isn't my first rodeo in the sports prediction business either. I also predicted that the 2001 Mavericks would make the playoffs for the first time in 10 years and that the 2007 Mavericks would have the best record in the NBA. I have witnesses.

Moving on -- I want to get my dork on and discuss web browsers. With Firefox 4 and Internet Explorer 9 in beta, I am actually excited about the evolution of web browsers going forward for the first time in a very long while. We have come a remarkably long way from the dark days where 90% of the world used Internet Explorer, despite its substantial technical inferiority to just about every other browser on the market.

I have decided to put together a small history of the web browser because I believe the current era of browsers is unique in the short history of the world wide web. Without further adieu, here we go, in awesome, awkward present tense bullet points:

The Netscape Era -- 1994-1998
- Netscape Navigator takes out NCSA Mosaic with innovations like javascript and frames. In 1996, Netscape reaches a dominant market share of nearly 80% at which point innovation stagnates.

The Dark Internet Explorer Era -- 1999-2004
- Microsoft, sensing a threat from Netscape, pours a bunch of money into the development of Internet Explorer. They come up with a few new ideas, but generally just copy Netscape and gain market share by bundling IE with Windows.
- IE eventually grabs 90%+ market share and Microsoft, resting on its laurels, disbands the Internet Explorer development team, effectively putting a halt to all browser innovation for years
- Netscape's browser undergoes a complete rewrite and its new codebase is released as open source. The Mozilla Foundation is formed to manage the open source project. Mozilla begins releasing a new product called "Mozilla Application Suite" which was modeled after Netscape's old Communicator all-in-one browser/email/composer product. Netscape releases versions 6 and 7 based on the Mozilla suite, with version 6 famously based on a pre-release quality version of Mozilla (version 0.6). The public, expecting one last hurrah from Netscape, is disappointed with the pre-release quality of Netscape 6. Netscape 7 is then released based on Mozilla's 1.0 version, but the public takes little notice. Mozilla, meanwhile, continues developing its rendering engine under the radar and makes regular releases of the Mozilla Suite, none of which garner much attention -- that is, until Mozilla changes its tack by creating a new, light-weight browser called Firefox.

The Firefox Era -- 2005-2008
- Firefox 1.0 is released on November 9, 2004 and, due to the fact that IE innovation is virtually non-existent, quickly and easily establishes itself as a vastly superior product. Tech savvy people take notice and Firefox eventually gains 20-25% market share. Despite Mozilla's efforts, IE continues to dominate, albeit to a lesser degree, by virtue of Microsoft's operating system monopoly, but is so clearly inferior that Microsoft engineers even admit as much publicly.
- After reaching ~25% market share, Mozilla's innovation machine seems to run out of steam. Microsoft releases versions 7 and 8 of Internet Explorer, but these are so feeble (especially version 7) that Firefox's superiority isn't really threatened despite Mozilla's relative stagnation.
- Apple launches Safari, but doesn't make much of a splash in terms of market share or innovation. (This is not to say that Safari was bad -- it just wasn't terribly original.) Actually, Safari was released in 2003, but the Windows version didn't launch until 2007. To this day, Safari's primary innovation was acceptable Mac OS X performance (which IE and Firefox did not provide at the time). The Windows version of Safari has never brought much of anything to the table, except for its WebKit rendering engine which became the foundation for Google Chrome.

The Chrome Era -- 2009 - ?
- Google releases Chrome on December 11, 2008, bringing major innovations with their new UI design and super-fast javascript engine. While Chrome's market share doesn't exactly explode, Google is extremely aggressive in Chrome's development. Meanwhile, unlike previous times in history, the existing browsers don't appear to be willing to let Chrome win without a fight. Mozilla and Microsoft are both beta testing new major releases of Firefox and Internet Explorer, each of which takes several of Chrome's ideas and attempts to improve on them. Apple released Safari 5 in June and is probably secretly preparing another release to keep pace.

This is really the first time that there has been more than 1 major company (i.e. not counting Opera) doing the innovating. In the Netscape era, only Netscape innovated. In the Internet Explorer era, nobody innovated. In the Firefox era, only Mozilla innovated. Now, however, Google, Apple, Microsoft and Mozilla are all innovating which is pretty exciting. Competition is definitely the driving force behind innovation. If you think about it, it's really amazing that it's taken over 15 years for us to have a healthy, competitive web browser market.

Finally, I would be remiss to discuss the history of the web browser without mentioning Opera. Opera has been a major innovating force in the background during all of the abovementioned eras. Version 1.0 of Opera was released in 1994 around the same time Netscape was founded. Many of the new features credited to mainstream browsers were actually lifted from Opera -- most notably tabbed browsing which most people probably assume was Mozilla's idea. There have been several times since 1994 when Opera arguably had the best browser, but it never really gained much adoption, in large part because they tried to charge for it while IE was free. Opera faces a pretty tall task going forward. In 2001, it was easy to create the best web browser, but now Opera has to compete with the three biggest names in tech (Google, Microsoft, and Apple) all of whom are putting major resources into their respective browsers, plus Mozilla which continues to be the standard-bearer of the free software/open source movement. Despite this onslaught, the latest version of Opera is fairly competitive with the mainstream browsers. It will be interesting to see if they can keep up.

Wednesday, September 16, 2009

2009 Rangers (nearly) postmortem

The recent dive by the Rangers likely ends their playoff hopes after a season that has been a breath of fresh air for beleaguered Ranger fans. The plan to develop from within that management committed to a few years ago has really started to pay dividends. The future looks bright.

After all I've seen this year, I'm not going to totally count them out, but, given that young pitchers frequently don't have the stamina to pitch late into September, the prospects are looking pretty grim. So, with that said, here is my review of some of the important players on the team going forward.

Starting Pitchers

Kevin Millwood - What started as a great season for Millwood is flaming out in spectacular fashion. Sabermetricians from all corners predicted a "regression to the mean" for Millwood after the All Star break. His knack for leaving runners on base and ultra-low BABIP (batting average on balls in play) were unsustainable given his career norms. But, instead of a regression, we've gotten an all-out meltdown which, honestly, came as a surprise to me. The guy looks like he ought to be a workhorse, but clearly he's pretty fragile. The team may consider shutting him down, but if they do, they will inspire conspiracy theorists to say that MLB is forcing the Rangers' hand to avoid his $12 million guaranteed option if he reaches 180 IP. Should be interesting, anyway.

Scott Feldman - You think Feldman has a shot at making the rotation out of Spring Training next year? After two years starting the season in the bullpen, Feldman has really established himself as a solid starter. A couple months ago, Bob Sturm pointed out that Feldman's extremely low strikeout rates probably indicated that he will come crashing down to earth at some point, but since that time, he has boosted his strikeouts, including an amazing 11-strikeout performance against Tampa Bay in August. Perhaps Sturm overlooked the fact that Feldman is still evolving as a pitcher. His new cut fastball continues to be the best in baseball (click on the wCT column header twice to sort), but isn't a strikeout pitch (though it does break a ridiculous number of bats). In his latest good stretch, Feldman has started to mix in his other secondary pitches (curveball and changeup) and the strikeouts have risen.

Tommy Hunter - A workhorse in the making, Hunter has been very, very good since being called up. I forsee a 200-inning machine for many years to come. Earlier this year, I heard people comparing Hunter to Rick Helling. That would certainly be nice, but Hunter's numbers this year have been better than anything Helling ever did -- and that's pretty exciting.

Derek Holland - Holland showed a few flashes of brilliance before appearing to run out of steam. He has gotten his feet wet and I expect really big things in the coming years.

Neftali Feliz - Feliz rose through the minors as a starter, so you know the Rangers will eventually try him in the rotation. Whether that happens next year, or the year after, I can't say, but it will happen. At any rate, if he ends up as a super-reliever or a starter, I will make every effort to be watching every time he's on the mound. What a talent!

Brandon McCarthy, Matt Harrison, Dustin Nippert, Eric Hurley - Depth for the rotation. All of those guys have a chance to become quality MLB starters and they will provide a nice insurance policy for the Rangers next year.

Martin Perez - The 18-year-old super-prospect has finished his season in AA Frisco. He was inconsistent after his promotion to AA, but that is to be expected given that he's the youngest player in that league by fully two years. There is an outside chance that Perez could make an appearance in Arlington next year, but we probably won't see him until 2011. His bona fides are as good or better than Feliz and Holland.

Bullpen

Frank Francisco - Injuries slowed what has otherwise been a very good season. If Frankie can stay healthy for an entire season, he could be one the best closers in baseball. (He could help his cause by eating fewer pies in the offseason)

C.J. Wilson - Except for a handful of bad outings, Wilson has been very good this year. He is most useful in a set-up role because he can go multiple innings. (Wilson used to be a starter) If Francisco stays healthy, Wilson can man the 8th and sometimes the 7th as well.

Darren O'Day - Best waiver signing ever.

The rest - Grilli, Guardado, Mathis, Strop, and a few of the starters from above round up the bullpen and they've been pretty good as well. Juaquin Benoit could be in the mix next year, coming off rotator cuff surgery. Guardado will probably retire, so they'll need to track down another lefty from somewhere.

Hitting

2 - Catcher - Salty held this position for most of the year until he was sidelined with thoracic outlet syndrome (a rare condition unless you play for the Rangers). Salty's defense was adequate (in other words, much improved) and would have been fine if he could hit anything. His line of .233/.290/.371 (BA/OBP/Slugging) is abysmal, especially for a player whose strength was supposedly hitting. After Salty went down, we got a pretty good dose of Taylor Teagarden. Teagarden's defense was inconsistent which is disappointing, given his scouting reports. (He was compared to Pudge defensively in the minors.) Meanwhile, though he probably deserved more at bats to get a rhythm going, his offensive numbers (.212/.273/.376) were actually worse than Salty. So, a position that was supposed to be a strength for the Rangers has turned out to be a major weakness and that leaves the Rangers with some decisions for next year.


The Greatest Catcher of All Time
Happily, the Greatest Catcher of All Time is also on the roster and looks to be a viable option for next season. I say we start Pudge next season and let Salty try to straighten himself out in AAA. It worked for Chris Davis this year, maybe the AAA hitting coach can work his magic on Salty too.

3 - First Base - It looks like Davis has mostly solved his issues from beginning of the season. If he has, he could hold down the fort at first base for a very long time -- especially given his outstanding defense. However, hovering just below in AAA, the Ranger's top hitting prospect Justin Smoak (also a first baseman) figures to be ready for the majors next year. Davis is probably the better defender, but Smoak is no slouch. I expect those two to alternate between 1B and DH much of next year.

DH/1B Hank Blalock's once promising career as a Ranger will almost certainly come to an end in a few weeks.

4 - Second Base - Kinsler had a bit of a down year offensively, though it does seem a little ridiculous to complain when your second baseman has a 30/30 season. Kinsler's defense has been outstanding and he's really cut down on his errors. It helps to have a superior shortstop as a double play partner.

5 - Third Base - Michael Young bounced back from a down year (playing with two broken fingers) with a tour de force season offensively (.322/.375/.523). He was consistently the Rangers best hitter all year long and, with two more homers, he will match his career high. For a player that appeared to be in a decline, it was a pleasant surprise. Defensively, Young is the same as always. He has limited range, but is very sure-handed when he gets to the ball. Overall, the fielding metrics rate Young as a well below average third baseman, but I expect him to improve with more experience at the position. Besides, we aren't in great need for a standout fielding third baseman because we have...

6 - Shortstop - Elvis Andrus has been a revelation at shortstop. Only Jack Wilson has a better UZR rating (a stat that measures fielding -- click on the UZR heading twice to sort) than Andrus. When Andrus cuts down on his errors (and he will with experience) he ought to be the best fielding shortstop in baseball or close to it for years to come. Andrus is the single biggest reason that the team's pitching has improved so much from past seasons. His hitting (.275/.338/.391) has outpaced most of the predictions I heard before the season and he is very much in contention for the AL Rookie of the Year. At 21 years old, his hitting will only improve. As I've heard some commentators say, if Andrus had just played this season in New York, everyone would be calling him the next Derek Jeter. I think we may have just witnessed the beginning of a hall of fame-caliber career.

Utility Infielder - The old man, Omar Vizquel has performed admirably this season. Fangraphs.com show his defense at 3B and SS to be excellent, and about MLB average at 2B. The numbers are very small samples, but I think they reflect a very strong performance in the field from the future Hall of Famer. Meanwhile, at the plate Vizquel rebounded from a very poor season last year and posted a respectable .273/.324/.354 (right in line with his career average, amazingly). All of that, and he doubles as a personal coach for Elvis Andrus; Vizquel has more than carried his weight this season. If he wants to play next year, I think the Rangers will welcome him back without hesitation despite the fact that he will turn 43 in 2010.

7-8-9 - Outfield
David Murphy - Murphy had a solid season, bouncing back from an 0-23 start. However, the Rangers outfield is so deep that he may be squeezed out next season.

Marlon Byrd - Byrd is a valuable player because he is so flexible. He can play all three outfield positions and he can hit almost anywhere in the lineup. Going forward, I see him as a super-4th outfielder. It will be more difficult to fit him in, however, if Justin Smoak is manning DH.

Nelson Cruz - After two aborted attempts in the majors, Cruz finally broke through in a big, big way this year. After 2006 and 2007, Cruz looked to be the epitome of a "AAAA" player. He scorched AAA pitching, but couldn't seem to hit in the majors. In 2008, the Rangers were naturally skeptical of his strong performance in AAA, but eventually his numbers were too overwhelming (.342/.429/.695) to ignore. When he was promoted to the majors, he finally broke through (.330/.421/.609) against major league pitching over 100+ at bats. This year, as pitchers adjusted to him, his numbers weren't quite as ridiculous as last (if they were, he would be Albert Pujols), but they have been respectable (.268/.342/.547, 32 HRs and counting). And, as an added bonus, his UZR has been the best in the majors among right fielders. Next year, Ron Washington needs to pencil Nelson Cruz at the clean-up spot every time he plays.

Julio Borbon - Borbon is the Rangers lead-off hitter and center fielder of the future. He has given every indication that he is ready take on that role right now. For some reason, the Rangers have been hesitant to play him in center field -- perhaps to let him concentrate on hitting -- but that should change next year. Borbon's elite speed will give opposing teams fits when he gets on base.

Josh Hamilton - And, of course, the biggest wildcard of them all: Hamilton. He set baseball on fire last year with an otherworldly first half, topped off with his fireworks display at the homerun derby in Yankee Stadium. Since then, however, he hasn't been nearly as good. This year, he was one of the Ranger's weakest hitters. Part of me wants to write off Hamilton's 2009 season due to injuries, and pencil the 2008 version of Josh Hamilton in the 3rd spot in the lineup next year. The Rangers, of course, will hedge their bets. That, more than any other reason, may be why David Murphy is a Ranger next year -- as an insurance policy.

Outlook for 2010

Barring a miraculous comeback, this Rangers team appears to be headed to a 9th consecutive season without a playoff appearance. It appears, however, that they have very good chance to stop that streak before it hits 10. The minor league system that produced Cruz, Borbon, Holland, Feliz, Andrus, Hunter, et al. is locked and loaded to produce several more waves in the next few seasons -- especially in terms of pitching. The aforementioned Martin Perez highlights a new crop of young arms that will continue pushing the Rangers into a new era where pitching is a strength instead of a laughingstock as it has been for a very long time. I, for one, am eagerly anticipating April 2010.

Monday, August 31, 2009

Health Care

Below is a response to a Facebook thread. I ran out of space in my comment on Facebook, so I went to a place where I am free to write as much as I want. :) The original thread is here and it references an article from the Washington Post here.

Now, without further adieu, my ramblings:

Small industrialized nations? The article references Japan, Germany, France, Britain which are the Nos. 2, 4, 5, and 6 overall in GDP around the world.

Why does our larger land area affect anything? The hospitals are already in place all over the country. We're just trying to figure out how to pay the doctors.

The fact is that the health insurance industry in this country is woefully inefficient. Around 20 cents of every dollar paid in premiums does not go towards paying medical expenses. That's huge -- the article mentions that other countries are able to get that number between 1.5 and 2. Our insurance industry employs armies of "adjusters" whose job it is to deny claims based on technicalities and loopholes. In other countries, where insurance companies are required to pay claims submitted by doctors, those adjusters are not necessary. By properly regulating the insurance industry, we could eliminate that substantial cost. However, if the insurance companies have to pay every claim, obviously we need to lower our medical costs as well. Perhaps we can't get them as low as Japan (who spends less than half per capita than us), but, as the article's MRI example illustrates, there is no reason we can't implement $98 MRIs here too.

It is a myth that government is inherently more bloated than private industry. Governments all over the world are able to provide insurance at a tiny fraction of the cost that our private industry provides here. Indeed, our own government provides cheaper insurance than the private companies with Medicare. The Democrats' plan (among other things) essentially proposes to expand Medicare to the whole population (the "public option"). The private companies will be forced to compete, but that's a good thing as far as I'm concerned. They're an oligopoly, so there's hardly any competition now.

But that's the plan that's on the table. Frankly, I don't think it goes far enough. Sadly, it is debatable whether even this timid plan is politically feasible. In an ideal world, we would have a single-payer system. (e.g. Medicare for all; no private insurers)

In my opinion profits made from health insurance are inherently immoral. They are literally profiting from the medical expenses of American citizens. Furthermore, they increase their profits by denying the claims of their paying customers.

Shared risk -- that's all insurance is. Why does that need to be a multi-billion dollar, for-profit business? What's wrong with the government handling that job? Clearly, it is capable of it (Medicare). The people who deserve to make money in the medical industry are the doctors and the drug companies. That's where the innovation and talent is. What does the insurance industry bring to the table? As far as I can tell, they perform a task that is better suited for the government while profiting substantially and denying millions of Americans access to health care.

This country isn't so different from the rest of the world that we are the only ones that can't have universal health care. It would certainly cut into the profits of drug and insurance companies, but it would prevent people like Kelly from owing Baylor Hospital tens of thousands of dollars for a 10-day hospital stay. If they had been on vacation in London when Lilly got sick, they would have gotten the same care and they wouldn't owe a dime.

Americans citizens are sick without access to doctors. They can't afford their prescription drugs. They hesitate before going to the emergency room because they're scared it might bankrupt them. Who are we? No other industrialized country does this to its citizens. If anything, we have a moral obligation to fix this problem, and the only thing standing in our way is the entrenched, super-wealthy health insurance industry which is the beneficiary of our misfortune.

Thursday, July 9, 2009

Lipstick on a pig...

I am becoming increasingly frustrated with the guy I supported and voted for last year. You see, I supported Barack Obama with the impression that he wanted to overturn Bush's most egregious policies. But Bush policies, even when administered by Barack Obama are nothing more than, well, lipstick on a pig as John McCain might say.


The Obama Justice Department retains its power of extrodinary rendition. It continues the policy of show trials. (Only prisoners for whom the government believes it can obtain convictions get trials). It even claims the authority to overturn the result of any trial if it doesn't go the government's way.


Read Glenn Greenwald for the juicy details. (here and here)

Mr. President, this is unacceptable. Please stop.

Not another dead blog...

And you thought I had quit blogging. I was just resting.

My friend Kelly alerted me earlier today about a new technology from Apple called HTTP Live Streaming. The linked AppleInsider article goes into great detail about the history of video streaming over the tubes, but I want to focus on a paragraph towards the end of their piece:

What's next? The obvious followup is to add support for HTTP Live Streaming in Apple TV, allowing for HD streams direct from broadcasters, facilitating the ability to only pay for channels you want to watch, skipping around the local cable monopoly while gaining access to content they don't carry.
That's right. HTTP Live Streaming + Apple TV = Over-the-Internet broadcasting. It could totally replace cable and satellite companies. I'm not saying that it will -- those companies will do anything and everything in their power to prevent that from happening. I'm just saying that the technology is there.

AppleInsider floats the idea of broadcasters (e.g. ESPN or MSNBC) bypassing traditional broadcasting vehicles and launching a HTTP Live Streaming feed. Anybody with a web browser could watch the channel in full HD. Anybody with an AppleTV (or any other HTTP Live Streaming-enabled device) could watch the channel on their existing television set -- no cable companies necessary.

More significantly, it would remove the obstacles of entry for startup TV stations. Under the current system, you have to have either a ton of cash or some really serious connections to start a TV station. Time Warner Cable et al. hold the keys to the empire -- you have to somehow convince them to put your TV station onto their channel lineup. But anybody can throw up a web server and broadcast via HTTP Live Streaming. It cuts out the middle man.

Al Gore had a really good chapter related to this in his book The Assault on Reason. (An excellent, excellent book which I very highly recommend) He talks about how our democracy was healthier before television because the news used to be more interactive. When the country was founded, the major news organizations consisted mainly of people with access to printing presses. You had stuff like Harpers Weekly, and less formal fliers etc. When newspapers came around, they were a little more removed from the readers, but still allow for a fairly high level of interaction. Heck, even I had a letter to the editor published in the New York Times one time.

Television, however, is totally non-interactive. There is virtually no way to contact a TV personality. It is one-way communication, and we're not involved at all. Also, TV triggers an instinctual reaction in humans that makes you focus your eyes on a moving picture. That is a survival instinct in the wild, but TV uses it to grab your attention and not let up. It triggers that instinct on every frame -- 30 times per second. In addition, TV tends to put your brain in a suggestive state, making you more susceptible to believe what you're hearing than you would otherwise. Reading, conversely, engages the brain and makes you more prone to higher level thought and critical thinking. Obviously, that makes TV a perfect tool to implement an Orwellian state and to kill democracy.

As long as TV is the primary way Americans get their news, democracy is in trouble. The vast majority of our nation's discourse is controlled by a very small number of people. That is why Gore founded his TV station "Current TV" which broadcasts viewer-submitted content.

Ultimately, the true hope for democracy, Gore says, is the Internet. It is far more interactive than anything before it with virtually no barriers to entry (anybody can start a blog). HTTP Live Streaming is just the kind of technology that the Internet can use to start whittling away at the traditional TV paradigm. Think how different the world would be if anybody could start their own TV station.

Let's just say that I don't think that it's a coincidence that Al Gore is on Apple's Board of Directors right now.

Thursday, September 25, 2008

Whoo hoo!™

Hm... I saw both of the following while I was reading on the Times today.

Whoo hoo™ indeed...



I've been kicking around a blog post on all of this for the past few days. So, consider yourselves (all both of you) warned. I may actually post something on this blog every now and then.