|
|
|
January 11th, 2009
09:42 pm - A Question
Since I have a "smart phone" (a term that is rapidly dying, and being replaced just with "phone"; maybe I should clarify as "internet-enabled phone"), does this mean that I should use Twitter? It would probably mean I would blog less (if that were possible), but I could get some syndication set up and make it look like I'm actually saying more.
|
December 28th, 2008
09:38 pm - The War on Apartment Clutter
So my new desk is (for all practical purposes) assembled. There's a piece that's supposed to attach to the top that hangs down and (as I later discovered) covers the gap between the drawers and the top, but it's "cosmetic", and I can hopefully attach it later this week. Pictures coming soon, once I figure out how to get them from my phone to Picasa (or a less Google-owned photo-hosting site).
I have to describe it as "low-quality", as at least two pieces had holes in the wrong spots, and a plastic "compression dowel" (like a regular dowel, only load-supporting) snapped in half because the construction directions given required it to hold more weight than it could at one point in the assembly. A nail (not included) fixed that, so no harm done. In general, the directions were not very good; I re-arranged a few steps during the process, and would do more now.
And with the desk comes two welcome tidings:
1) The old, old (my parents may have bought it before I was born) folding card table is now in the dumpster.
2) The large pile of bills, credit card statements, pay stubs, etc. that I don't really care about because it's all online is gone (I guess it's a small pile of my 2007 taxes and my car insurance papers). The other papers are now in the trash (envelopes/ads), my bag (to get shredded), or the file drawers.
The next step is (most likely) moving stuff around in my apartment. I'm not sure if it makes sense to have my desk in the more-desk shaped area by the kitchen, and to move the kitchen table more into the living room area. I'll figure something out. Also, I should start up my desktop and probably start the decommissioning process. It's 6+ years old, I rarely use it, the monitor stand broke so the monitor is at a terrible angle, and it takes a lot of desk space. I've got lots of old stuff on there I may want to keep, though.
And then the next project will likely be that my book collection has overgrown my bookshelf and I need another one somewhere. I'm at about 3.9 / 5 shelves used, so I probably have about 3 months at the current rate before it starts to seem full.
|
November 30th, 2008
08:24 pm - Thoughts on the Phone
I'm back to my apartment, and before the maelstrom of Google starts to re-envelop me, here's a review of my new Google phone (aka Android, aka G1).
Like: Mobile-optimized versions of websites. I'm talking about you, Google and Facebook. Both have homepages highly optimized for mobile screens, with less information on the page and more links useful on mobiles (Goodbye links to Youtube and Maps; most phones have separate apps for these and no Flash).
Don't Like: Regular sites on the phone. Besides Yahoo! and Microsoft, neither of which have done anything for mobile, sites like cbs.sportsline.com and cnn.com also have issues with usability. Assumptions of at least 640x480 generally break down, and elements such as SELECT boxes and iframes work (or don't work) completely differently.
Like: Dedicated apps for sites. As noted above, usability can be a pain, so if you can't do what you want in a browser, write an app for it. I would love a good "sports scores" or "stock quotes" or "TV schedule" app, even though I know sites that do this.
Don't Like: Lack of apps. Most of the apps listed exit for the iPhone, but not Android (at least not yet). This should pick up on the next few months.
Don't Like: Mis-clicking. It's very easy to click in the wrong place or hit the trackball too much. This can cause havoc by loading links you don't want to load, or doing the wrong move in a game.
Don't Like: The unbalanced keyboard. As I noted, typing with the two thumbs is painful on my right fingers, since there's the part of the phone where the "home" and "end" keys are on the right side, causing your fingers to have to stretch. Alternate typing techniques should help.
Like: Integration with Google Contacts. This is great from a Google strategic point of view (more people using contacts = more people using GMail), but it's also nice because emails and phone numbers are now really tied together.
Like: Free ring tones. This is the type of thing where having an "open" system will help, it's trivially easy to add ring tones or create your own from existing songs (Ringdroid is an app to do exactly that). I sense that the "ring tone" market is nearly dead, and with it the stupid commercials to spend $20 bucks for 10 ring tones per month.
Like: "Draw pattern to unlock". Not less secure than a password, but easier to remember and faster to do. Also serves well to prevent accidental use in a pocket, which is the main point.
Don't Like: Lock speed. Phone will go to sleep mode (good) and lock itself (less good) after about 30 seconds. If it didn't require a pattern for the next minute (perhaps "click here" to prevent accidental use) it would be nicer.
Like: Maps and Youtube apps. They "just work". Very good.
Don't Like: Camera app/camera. You have to hold still for about 5 seconds to take a non-blurry shot. Even if I don't shake, I often assume the picture is taken and move before it actually takes.
Like: MicroSD card, particularly after upgrading to 4GB. Enough to keep enough music as well as other stuff. Very tiny, actually.
PS: Like: Automatic notification on email arrival. I get a volume of email at my public account (5 per day or so, most of which I care about) that this is very useful.
|
January 1st, 2008
12:00 am - The Parting of the Ways
Well, after 1330 days, 1225 posts (counting this one), 1270 received comments, and a bunch of other statistics I don't want to bother to compute, we have reached a crossing roads for this blog. It's being sharded three ways, as of today.
1) The "link blog" lives at Google Reader, containing all the day's news stories that are interesting in some way. If there were an easy way for me to comment on them and I had more time, I would.
2) I'm starting a blogspot blog for "serious" posts. To start with, this will probably be mostly about technology, politics, and a bit of random other stuff that seems serious. It specifically will not comment on anything related to Google. It also will not talk about anything related to who I am or what I've done recently. The url is http://appower.blogspot.com/ . This blog, in line with the theme established here, will be named "Pictures at an Exhibition". Expect volume of at least 2 posts per week to start with.
3) This blog isn't going anywhere. Er, that's not exactly true. It's going in two directions. First, I'm going to try to give more detailed and frequent posts in the "what's going on in my life" direction, that aren't just random ramblings. Second, it's going friends-only after this post. I'm pretty sure that most people who read this have a large number of their LJ posts as friends-only already, so it shouldn't really matter. For those of you who might be reading this some other way, I will syndicate some of the posts to Facebook notes, which should be readable by you. If you are neither an LJ-friend nor a Facebook friend (my profile), and you can't bring yourself to ask for either, then you're just out of luck.
|
December 28th, 2007
09:50 pm - The Domain Registry
It costs a domain registrar $6.62 in fees to ICANN and Verisign to register a domain name. This is around 1.8 cents per day. With a well-designed hosting system, it probably costs not much more than that to host domains in bulk as well, and set up some auto-generated content with lots of ads. If you can get 2 or 3 people a month to come to the site and click on ads, it's a profit-generating business. And with billions of people on the internet, many of whom are keen to make typos and click on the oddest ads, that doesn't take that much effort. It would be relatively easy for one person to scale this to 10000 domains with a dozen or so machines, a dedicated internet connection, and a few simple scripts that could be written in around a week.
Thus, people do this. A lot. Some of them make money, many of them don't. Regardless, it leads to millions and millions of "parked domains", where people spend the $6.62/year (probably around $7 with the registrar's commission) to hold onto the domain and generate the trickle of clicks. At this point, in the .com realm (which is still the only one that matters for any business), probably every 3 letter domain is registered, as is every 4-letter alphabetic domain and every 4 or 5 character numeric domain (you can get random 4-letter hybrids, like d3gf.com, z3gf.com, hea4.com were some randomly available ones). Any logical English word is registered, as are most logical combinations of words.
In one sense, this price is fair, since ICANN and Verisign really don't have that much in the way of costs per domain. They have to have a large replicated database and a lot of serving capacity, but each record is so small it amortizes to maybe 15 cents per record. On the other hand, demand for many domains is a lot greater than supply. So there is a secondary market with utterly obscene prices (For example, Computers.com for 2.2 MILLION dollars, CaribbeanVacations.com for 130 thousand, News.mobi for 110 thousand, RapVideos.com for 70 thousand, CertifiedPublicAccountants.com for 30 thousand, DrugRehabilitationProgram.com for 8 thousand). What we have here is a land grab, and lots of people hoarding domains and doing nothing with them. It also looks suspiciously like a Dutch Tulip auction, but that's just me. Ideally, there would be something akin to property taxes on domains or some other way to drive the price up to prevent people from registering them and doing nothing, and causing startups to have stupid names like Meebo and Kijiji and Foonz.
However, the single letter domains mostly aren't registered to anyone. Paypal owns http://www.x.com/, Nissan owns http://www.z.com/, and Qwest owns http://www.q.com/ . In this blog, some guy suggests that ICANN auction off the domains to raise money. Fair enough idea, even if the auction scheme sounds like it is ripe for collusion. However, the comments are priceless in their stupidity. They're worse than Youtube. And, if you replace ICANN with "the government", pretty clearly show flaws in conservative logic that conservatives are more careful to cover up. ( This is basically shooting fish in a barrel ... some of the comments )
|
December 27th, 2007
08:02 pm - Good Morning, DMV
Random Fact of the Day: If you drop a half-full glass bottle of Orangina from 2 and a half feet onto a parking lot, it won't break! At least mine didn't.
Random Fact of the Day Number 2: Firefox spell-checker doesn't like the word Orangina. This puts the all-time score at me: 258, spellchecker: 3. (259, since I think it's perfectly fine to make spellchecker into a compound word)
So, despite going to the DMV in early October to get my driver's license, I am yet to actually RECEIVE it. This is because California has a bizarre policy where you don't actually get the license when you are at the DMV, you have to wait for them to mail it to you, supposedly within 60 days.
Now, it's been significantly over 60 days, so I called the DMV to find out whether they were being slow, or if (possibly) it was the post office's fault. I call them up, and I get a minute-long recording "The DMV will be closed on Dec 25 and Jan 1" in English and Spanish and "Try using our website!" in English and Spanish before they give you any options. "For English, press 1". *1*
Another minute of talking, letting me know about the changed hours, and how I can do some things online now. "To check on driver's license, press 2". *2*
"Please be aware that it may take up to 60 days. If it has been 60 days, press 0 to talk to a person." *0* *ring, ring*
At this point, I get a person who conveniently tells me after a minute of getting my info that I need to call a different DMV number to find out what is going on, and ask about a PDPS inquiry.
I dial. Busy signal. 5 minutes later, I dial again. *ring, ring*
The second person, after entering my information, lets me know that they sent for a PDPS inquiry (something about if you have points in another state) about 60 days ago. They normally come back after 30 days, but mine apparently hasn't. He told me I would have to call a third number to check on the status of that. Also, I could go down to the DMV to get another temporary permit. Considering I'd have to wait an hour, I think I'll pass.
I dial the third number. Busy signal. 5 minutes later, busy signal. I get distracted and don't get back to calling today.
In conclusion: I think a person could get elected Governor of California by running on a platform of reforming the DMV.
|
December 22nd, 2007
11:46 pm - Top 10 RSS Feeds
By popular demand (er, maybe not so popular, but who knows), I have compiled a "powera's Top 10 Feeds" list!
Dishonorable Mention:
http://www.blogsforvictory.com/ - Wait, this isn't "Bottom 10"? Still, Mark Noonan is the best "right-wing moron" on the entirety of the internets, which is some form of accomplishment, I guess.
http://precursorblog.com/ - On the same "Bottom 10" line, Scott Cleland is the best "I hate Google, Moveon, and anything else possibly hostile to telecom companies" moron on the web.
Runners-Up (there are 10 of them):
http://www.basicinstructions.net - Pretty good webcomic, but it gets a tad formulaic over time
http://iowahawk.typepad.com - Iowahawk, the only one of the conservative political humor commentators that's ever funny
http://www.allfacebook.com - If you want your Facebook news.
http://www.mattcutts.com/blog/ - If you want tips on Search Engine Optimization from deep inside Google.
http://kara.allthingsd.com/ - A good "tech industry" blog.
http://www.gigaom.com/ - A good Silicon Valley focused "tech industry" blog.
http://www.realityblurred.com/realitytv/ - For all your reality TV "news"
http://www.boingboing.net/ - It's Boing Boing. You probably already read it.
http://www.valleywag.com/ - It's the "National Inquirer" of Silicon Valley. Occasionally, they do break stuff. However, they're usually wrong.
http://www.scobleizer.com/ - Sorry Scoble, you just missed the cut. He's got lots of long videos. He's also not afraid to be wrong, which is both a good and a bad thing. He's got lots of good posts, but also an hour long video of how Mahalo is going to beat Google, which is just ridiculous.
And the winners:
10) http://blog.pmarca.com/ - Marc Andreessen's blog - Good, solid commentary on the tech industry mixed in with the world at large.
9) http://jwz.livejournal.com/ - JWZ - He used to be in the tech industry, and wrote stuff like Netscape and xscreensaver. Now he owns a nightclub, and writes about ... whatever.
8) http://www.joelonsoftware.com/ - Joel Spolsky - Lots and lots of great ideas in his posts. If he didn't try to make everything be somewhat of a promotion for FogBugz (a product I've never used with a terrible name), this blog would be much better.
7) http://www.nerdnirvana.org/ - Sangent: The Daily Ramblings - Lots of amusing YouTube and other videos. I'm particularly fond of the "The Real Hustle" videos he gets from the show in the UK.
6) http://strangemaps.wordpress.com/ - Strange Maps - Amusing historical maps. 'Nuff said.
5) http://www.xkcd.com/ - XKCD - Do I need to elaborate? It's XKCD.
4) http://www.politicalwire.com/ - Taegan Goddard's Political Wire - Good brief summaries of the day's most important political news, without too much of a political bias.
3) http://www.dailykos.com/ - You may protest his politics, but Kos is very good at both analysis and running a website. The presence of diaries mean that there are literally thousands of budding political writers there, and the best of them move to the front page. Former bloggers there populate a large portion of the liberal blogosphere, but a good number of the best still post on the front page. The group blog aspect also allows it to hit top news 24/7 with limited latency, and to have experts in many fields. There really isn't a conservative analog, since Free Republic had such an early start that it swallowed up any efforts on the right.
2) http://googlesystem.blogspot.com/ - Google Operating System, by Ionut Alex Chitu - The best blog on all things Google, period, hands down, end of discussion. I work at Google, and I read this blog to find out what is going on. It's not affiliated with Google, but it's great at finding all the new Google products and giving a reasonable analysis.
1) http://www.techdirt.com/ - The best tech blog I've seen. Lots of relevant posts, with thoughtful analysis. This has the best signal-to-noise ratio for a blog of it's volume in posts per day by far, and accordingly has the most shared posts in the past month, beating out blogs with 3 times as many posts.
You can get the Top 10 blogs conveniently at http://www.google.com/reader/shared/user/04455784389254111638/label/Top10 and the runners-up at http://www.google.com/reader/shared/user/04455784389254111638/label/RunnersUp .
|
11:43 pm - Oikonomos
"The children asked him if to kill was not a sin
"Not when he looked so fierce" his mommy butted in
If looks could kill it would have been us instead of him" - The Continuing Story Of Bungalow Bill
I like to cling to the fiction that prices are in some way directly related to the cost of production. In Economics 101, the model is that if prices are sufficiently above the cost of production, there will be competitors, and the price will naturally get knocked down.
However, that doesn't work for multiple reasons. First, you have patents, copyrights, and trademarks. Second, you have innovation that occurs faster than competitors can keep up. Third, there's implicit collusion on prices in some situations, since nobody wants a price war. Fourth,
There are two incorrect conclusions that can be drawn from this. First is the timeless "Just because it costs more means that it's better/more valuable". Second, we can get "By buying the cheapest thing that satisfies your needs, you are not only saving money, but also helping the store and the rest of the economy behave more efficiently."
I tend to lean towards two other fallacies as well: "Rent seeking is always bad and those doing so should be shunned", and "Waiting in line is generally bad, and a symptom that you should try at a different time". As for rent-seeking, the main issue is that it leads to people making (possibly large) amounts of money by basically being in the way, and it in general is not very useful. As for queueing, it also does serve as a way of effectively increasing prices for a scarce good. If the good will become less scarce in the future, it's generally a good thing to avoid the excess queueing. If the good will always be scarcer than demand, then, well, you're generally out of luck, and have to just try and get lucky.
|
December 19th, 2007
01:59 am - My General Schedule for the next 2-3 months
December 19-25 - My brother is in town visiting from Iowa. Sightseeing and tourist-y stuff abounds, probably including going to the 49ers-Tampa Bay game, and probably working on furniture and household accessory acquisition.
December 25-31 - Generally working and on-call, so the other people in the rotation who actually care about Christmas aren't. I think I won't be on-call for New Year's Party time (we're working to get Dublin to cover that in our timezone), but haven't figured that out for sure yet.
January - Nothing exciting scheduled yet for this month. If you're in the SF area or know stuff that is, feel free to suggest stuff.
February 4-6 - Valleywag was right, Google is going to Disneyland! Everyone in the western half of the US in the company is going to Disney on Tuesday, with 2 travel/party days book-ending it. I'm thinking that I may leave on Friday, February 1 to go to LA to sightsee, since I'm yet to be in that town. If you have thoughts as to why that would or would not be a good idea, or have ideas on what I should do when down there (apart from corporate-sponsored drunken debauchery), please comment.
February 8-11 - On a red-eye arriving Friday morning, I fly back to Iowa. My other brother's Bar Mitzvah is on Saturday. I still need to book a one-way flight back. Flights out of DSM are ~$100, flights out of CID are ~$300, and flights out of ALO are ridiculous. I may do some convoluted tour of the state to end up flying out of DSM on Wednesday, especially if ![[info]](http://l-stat.livejournal.com/img/userinfo.gif) deathweasel, who will be in CID, doesn't have a reliable car yet.
rest of February - Google doesn't run itself, you know. Somebody has to feed the gnomes.
Oh, I got business cards. They say "Alexander Power, Gnome Feeder, Site Reliability Engineering". See, I told you I wasn't lying when I said I feed the gnomes. It says so on my business cards even!
|
December 18th, 2007
10:58 pm - Iowa Caucus Analysis
Any analysis of the Iowa Caucuses really needs to keep in mind 3 things which may be different about the caucuses than a regular primary election:
1) Only people who physically attend the caucus at 7PM on Thursday, January 3, can participate in the caucus. There is no absentee ballot, no make-up date.
2) On the Democratic side, there is no secret ballot in the caucus. You vote by joining a preference group. Republicans use the results of a secret ballot.
3) On the Democratic side, only candidates with at least 15% of the vote in a precinct can get any votes from the precinct. Further, only committed delegate counts are announced. Republicans, once again, use a secret ballot without a threshold for viability.
This impacts the Democratic results a lot more than the Republicans, which is mostly a regular election apart from the "specific place and time" aspect.
In general, this means:
Dennis Kucinich should do exactly as well as he did last year, hitting viability in a few precincts by Fairfield (home of the Maharishi college), and in major cities. Gravel may fail to be viable in any precinct. Al Sharpton and Joe Lieberman didn't reach viability in 2004, and Wesley Clark got only .1%.
Chris Dodd, Joe Biden, and Bill Richardson are likely to have their support appear significantly lower than it would otherwise be from the viability constraints. I would expect a candidate with 5% support statewide to get at most 1% of the vote in the caucus. Considering that all endorsements and momentum is being gobbled up by Obama, Edwards, and Clinton, I don't see how any of them will break 2% statewide unless they can push closer to 10% in statewide polls.
I think the decision between Obama, Edwards, and Clinton is still too close to call at this point. I wouldn't want to speculate without being on the ground in Iowa. However, assuming that they each get 25%, and the win doesn't cascade into a larger win in New Hampshire, there probably will be nothing decided until the mega-primary on February 5, by which point at least one candidate should be practically eliminated. If it's close enough between the other two, they could keep running to probably force a brokered convention, but that's very unlikely (I'd estimate a 40% chance all 3 are viable for the Feb 5 primaries, and only about a 5% chance all 3 keep going after that). Edwards obviously has the most ground to make up, but the sheer math in Iowa looks good. I'd estimate the 2004 supporters to break down about as follows (this is a complete WAG):
Kerry - 60% Clinton, 30% Obama, 10% Edwards
Edwards - 70% Edwards, 20% Obama, 10% Clinton
Dean - 50% Obama, 30% Edwards, 20% Clinton
Gephardt - 50% Clinton, 30% Obama, 20% Edwards
which gives Clinton 35%, Edwards 34%, Obama 30%. This is obviously missing lots of factors and based on a lot of wild guessing, but it's my best attempt at a prediction. Overall, I'd put my money on Obama, since he seems to have the most upside left.
The Republicans are more interesting, with 6 candidates figuring to have a possibility of winning the race: Thompson, Romney, Giuliani, McCain, Huckabee, Paul. I have them listed in the order from "most establishment" to "most non-establishment".
The FReepers are generally Thompson supporters, which should say something, but he hasn't done a thing to distinguish himself other than being the great hope of the establishment powers that be. Romney is certainly the "big business" candidate, and the "I stand for everything the Republican party is under George Bush" candidate. Giuliani is the candidate of the fascist wing of the Republican party, and has various problems on "social issues", but "9/11" and "I can beat Hillary" and did I mention "9/11"?
McCain used to play the radical, but seems to have gotten too tired of being attacked by his own party to be too much of a disruptor, and is getting the lion's share of endorsements, from Joe Lieberman to the Des Moines Register. Huckabee frightens the big business part of the Republican party but has the Christian base and Chuck Norris. Paul frightens most all of the Republican party that doesn't support the gold standard and leaving Iraq, but has the most fervent supporters and (astonishingly) the most money, and should get >15% in New Hampshire regardless of what happens.
Then there's Duncan Hunter (supported mainly by some die-hard Freepers), and Tom Tancredo (still pushing against illegal immigration, legal immigration, and foreigners in general). Once again, they can be best ignored, except for their endorsement if they decide to drop out. Tancredo may go all the way to the convention on issues, but Hunter may endorse relatively early, perhaps before Iowa if he doesn't want to be crowded out by bigger fish or feels he can guarantee a theoretical cabinet slot.
Also note that a lot of states on the Republican side use winner-take-all for delegate allotment, which can easily cascade small margins in votes to large ones in delegates. (The Democrats banned this after the '68 convention) These states include (from The Green Papers): Arizona, Connecticut, Delaware, Missouri, New Jersey, New York, Utah, and Virginia, with many other states using winner-take-all by Congressional District (including California, Florida, Ohio, and Michigan).
If I had to predict overall, I'd be most wary of Huckabee, as Markos of Daily Kos has felt since May 2006. The only other feasible option I can imagine is Romney, as the Establishment candiate. The Powers That Be realize that Fred Thompson is a lazy old man who doesn't have a single interesting idea and has no charisma, and they won't prevent him from going nowhere. Giuliani is probably sunk regardless of what happens. McCain ... McCain is an enigma. I can't imagine he'll do well in Iowa, but if Huckabee wins Iowa it may end up McCain-Paul-Huckabee in New Hampshire, which could give him a fighting shot. Paul cannot win unless he manages a complete sweep on Feb 5 because of the super-delegates, but he could do well enough to split the party in half.
|
|
|
