Monday, December 27, 2010

The future of work

As I mentioned back in 2007, I believe that we will eventually enter an age of leisure, largely due to robots and technology. We are seeing this trend accelerate in the United States currently, as companies are finally beginning to see real cost-savings from technology. In the initial ramp-up, technology improvements employed a ton of engineers and programmers. In its later phases, these contractors will be out of work. Martin Ford has written a lot about this and his blog Future Economics and Technology keeps up on the developments in that direction.

It seems to me that this view is quickly becoming more common. A few days ago I saw What happens when we have enough technology that there is no need for a work force? on Reddit. There were some people who said this could never happen, such as "skeeto", but when pressed it was noted to him that the robots could be designed to be identical to humans (the Japanse are already making humanoid robots - see video). It's true that there will always be some demand for workers, but I think it's naive to think that more than a small percent of potential workers will be needed in the long-run. We'll have to figure out some way to keep the rest of people occupied, and we'll have to have a system which supports them.

I do not expect the unemployment rate in the United States to drop back to 5% in the next few years, and possibly never. I actually expect it to be in a slow uptrend, although we'll probably see a dip if the baby-boomers actually retire like they should. We might have to figure out a different economic system sooner than people think. The Reddit thread directed me to an interesting story called Manna which describes the possible future. It's not exactly as I see the future, but fundamentally a lot of it is plausible.

Friday, December 17, 2010

VBA and bankers' rounding

Most children learn in elementary school that you round 5 up. That's how Excel works. But the language used to write programs in Excel, Visual Basic for Applications, does not do that with its Round() function. It uses bankers' rounding, which is described by a Microsoft blogger. You can write a custom function such as one here, but that's kind of a hassle.

It's annoying, but there is a workaround: call the Excel Round function, as described by this guy.

If VBA was open-source, I imagine that this would have been fixed a while ago. Regular round would be Round(), and Bankers' round would be Bankround() or something. VBA seems like a mess.

Sunday, December 12, 2010

The Institute of Medicine's vitamin D report

I watched about 50 minutes of the IOM's video of the new report's press release. I thought it was remarkable how sincere the scientists were: they truly do not understand a lot of things going on with these vitamins, they're upping the vitamin D required significantly for most population groups from a 200 IU adequate intake to a 600 IU Recommended Dietary Intake with a 4,000 IU tolerable upper limit. They ignore vitamin D from sunlight.

They say later that most people in the United States get about 200-300 IUs from the diet. They grudgingly admit that most people are going to have to take vitamin D supplements; later they discuss that people could specifically target fatty fish (e.g. salmon) or specific fortified foods. When asked about what they think about people taking 1,000 or 2,000 IU supplements just in case, they said it probably wouldn't increase risk, although one cautioned against it noting the beta carotene and vitamin E examples, which looked good in observational trials but proved unhelpful or even harmful in randomized trials.

It's rather sad that they've done all these randomized controlled trials on bone health, but they didn't collect all this other information on these trial patients (cancer, immune system, etc). Without the randomized trials, the IOM could not say anything, which I don't blame them for given the history of observational trials. I'll be taking a 2,000 IU supplement to be on the safe side, and I'll be upping it from the occasional couple times a week to a routine thing. I'm still partial to the idea that the seasonal flu is caused by vitamin D deficiency.

If you aren't looking for something then you'll never find it, which is why I'm constantly disappointed by people who seem to reject the idea that focusing on the nutritional basics (rather than taking a bunch of drugs) can effectively improve health. It's also rather sad that we still don't even understand how something as basic as vitamin D relates to immune function when we've expended enormous amounts of money on various drugs which were harmful dead ends.

I may add more thoughts on this later.

Saturday, December 11, 2010

Interest rates, banking profits, and predatory lending

I like to get firsthand experience of things. When I heard that a local bank of mine was missing its TARP payments and possibly close to bankruptcy, I thought about opening an account. I wouldn't mind getting the experience of having my accounts transferred or being paid by the FDIC. Lately I've been lucky: I have an open directors & officers' insurance policy claim (I didn't file the lawsuit) and I recently got a $1,000 payout for a dink in my car (first insurance claim).

About a week ago I went into Wells Fargo to wire some money to China. I had purchased a credit report on a Chinese stock that I was invested in - the credit report used the SAIC (State Administration for Industry and Commerce) financials, which are often different from the financials that these companies file in the United States. The credit report confirmed my suspicions and I closed my investment at a loss. I was helped by a personal banker, and while I was there she suggested that I take out a personal line of credit.

I told her I didn't need it, but I was easy to sell on the idea: I wanted to see what the experience was, and I wanted to see what kind of interest rate they'd offer me. They found I have a credit score of 777 and I disclosed that I make about $45k a year. I have no debt. They offered me $6,600 with an interest rate of 8.750% plus the Prime Rate published in the Wall Street Journal Western Edition. As of November 29, 2010, that meant an interest rate of 12%. However, this was subject to a floor of 14.50% and included a discount of 25 basis points for automatically deducting payments from my checking account. I also went through a process of upgrading my checking account, and in that process I discovered in the documents provided that Wells Fargo actually does payday lending through a program called "Direct Deposit Advance".

I had reminded the banker several times that I was not interested if there were no fees. Reading the contract, I saw an annual fee of $25. I spoke to her about it and she said her manager assured her that it was waived the first year. So I went along with it. When I get the line, I was immediately charged $25, and interest began to be charged soon after ($0.03 was charged as interest). She offered to remove the fee, but I was glad to be rid of the line. I guess that goes to show that you can't trust words. If it had been serious, I would have gotten it in writing first, although I doubt that would have mattered much with the computer's programming.

Although not clearly related to the above story, I've sometimes idly wondered what the net effect the Fed lowering interest rates was on banks. The banks charge less in interest, but they also pay much less in deposits, so the effect on the net interest margin could be OK. As my experience shows, banks can still charge a fairly hefty interest rate to me, a borrower with very good credit. Plus, the lowered interest rates stimulates additional loans. The Dallas Fed's Southwest Economy said in 2003 that the lowered interest rates have, prior to 2003, been closely correlated with higher net interest margins, basically because low short-term interest rates are typically associated with a steep yield curve, and banks (accepting the risks of maturity mismatch) invest the deposits for which they are paying very little interest in longer-term assets. The effect had notably broken down in 2001-2003, and the author says this is because "interest income on many of [bank's] floating-rate loans falls with market rates, but deposit rates on short-term accounts fall by less as overnight rates get closer to zero and account management expenses become relatively more important". I don't exactly follow how this is different from the earlier years, but I guess it has something to do with the trend towards banks getting involved in money market funds.





A propos of the above discussion, Morningstar recently sent out an article on the implications of interest rate hikes (When Interest Rates Rise, Will the Tide Lift All Boats?). They target retail brokers as the greatest beneficiaries,

Wednesday, December 08, 2010

Accepting web design and funky CSS

I've made a few websites from scratch in notepad and played around with programs like Dreamweaver, but I was never very happy with the results and my understanding. The websites felt amateur, and CSS always seemed to do weird things. I gave up on further work on websites thinking that I was missing something about web design - that I wasn't quite up to the standards and secrets. In the meantime, I've collected a mess of links on web design which I thought someday might help me. HTML Dog and W3Schools are the two that I am most familiar with. In my quest to understand CSS, I'd also bookmarked thesitewizard.com, HTML Source, HTML Help from The Web Design Group, Good Tutorials which lists a bunch of other tutorials, an interesting extension of CSS called "HSS", and the Open Directory's listing on CSS. For general web design, I'd also found a list of 100+ resources, a handy "complete color chart". Suffice it to say that I did not read all these links carefully. It's the problem with bookmarks: it's easy to file away and forget. However, after all this I never took the time to look at the official source, W3's official specification (see the CSS 2.1 draft). I did look at it today, and found it somewhat helpful although still not completely clear. In my quest, I also bookmarked Wikipedia's Comparison of HTML editors; now, I also see another list of web design programs on AlternativeTo.net.

In my search, I found articles which I thought offered solutions, such as strange hackish One True Layout, and teaser samples of Sitepoint books such as HTML Utopia: Designing Without Tables, The Art & Science Of CSS or Everything You Know About CSS is Wrong I never bought those books. I also found that I wasn't the only one frustrated by CSS. In February 2009, I found an article by Eric Meyer (Wanted: Layout System), lamenting the fact that CSS has historically lacked any decent layout system. I also found a Reddit discussion (Does anyone else who doesn't work with CSS on a regular basis feel like a champ whenever they get it to work?). I didn't really understand what these posts were talking about at the time, and I still only vaguely follow them. For example, I don't understand Eric's point that the only reason floats are used is that there's the clear value which allows footers to be set at the bottom. I can use floats to create a decent layout without clearing at all. However, I agree with the general idea. When I think of web design, the tables based method is the most intuitive. I should be able to cut the screen into little pieces which don't overlap into each other. Floats are a hackish way to do that, but there should be something better.

One of the commenters to Eric's article mentioned Blueprint (and the similar program 960 Grid System as a way to get past this issue. I don't understand that. I get that Blueprint is a handy way to deal with defaults and special Internet Explorer quirks, but I don't understand how their grid makes CSS more intuitive and less hackish. Why split a webpage into 24 columns, and then use those 24 columns to create a basic layout with 3 columns, or however many columns? Working with the grid has its own rules (shown pretty well in the tutorial A Closer Look At the Blueprint CSS Framework). Further, it uses a fixed layout, and I generally come down on the side of using a relative "liquid" layout (see The Great Fixed Vs Relative Width Page Layout Debate).

After all this, I'm still rather confused. I think ultimately I need to get down to actually working on websites and hacking through problems however I have to. I don't actually need a perfect layout system. I can work with what's there, and what's really important in my website ideas are the databases and related programming. So I'll be trimming out my web design links and leaving instead the bookmarks on databases and programming. Perfectionism is the enemy of productivity.

Sunday, December 05, 2010

Programmed music?

The problem with bookmarks is that there's no schedule reminding you to look at them at a certain period or deal with them. I used to use them for interesting articles or stories that I'd read later, but instead they built up as I moved on without finishing what I'd bookmarked.

One of these stale bookmarks is a press release about a program called Inmamusys (for Intelligent Multiagent Music System) which can compose pleasant music on the fly. No use to me, but I thought it was cool and it promises great, non-repetitive elevator music. Of course I'd be wowed and completely believe this kind of hype...

Giving What We Can project

I added this as a bookmark while I considered joining the project, but I'm pretty firm at this point that I will not join it. I'm looking to give about 10% of my income away in various causes, but not all of them are charity-focused. When I've built a small fortune I plan to return to this.

Trimming a few blog and other bookmarks

As part of an effort to continue to trim my bookmarks of old articles and half-thought ambitions, I'm removing a few blogs from when I was more obsessed about health, plus an article ("knol") about eyeglasses possibly exacerbating myopia ("Advisory on Eyeglasses for Myopia"), the Building a Mining Company website, Mines and Communities website, and all my privacy websites (Guerrilla Mail for temporary websites, Bugmenot for free logins). Not sure what to think about the knol, and my interest in mining will continue but somewhat more from an investor-based perspective (where I can look at the ecological impacts as well). I also trimmed the only link in my funny folder: Motivational Posters.

The blog include the Heart Scan Blog, which is about a doctor who advocates vitamin D and other things (thyroid stuff?). Caught my eye, but I haven't read much of it lately. Another is Whole Health Source, who caught my eye with an interesting article on Weston Price's research on tooth decay. I actually agree with both of the bloggers to some extent, or at least in the idea that mainstream medicine is often wrong and that the importance of nutrition is overlooked. But since I don't have as many health issues as I used to, I'm less into this area. Another is Modern Dragons, which seems like a guy with a common interest in tackling "modern dragons". There's also The Scholarly Kitchen (academic publishing blog), This week in batteries (need more technical background to follow).

I'm adding the following to my Google Reader: Aaron Schwartz's Raw Thought, Andrew Garret's Blog, Wikimedia Technical Blog. These are programmers who have a connection to Wikipedia, although I'm trimming out Wikipedian blog Vox Populi Vox Cani. I'm not adding Vision Master Designs also I'll likely go back to check on it (same with all of these, now that I've got them stored here). It's unfortunate that Google Reader doesn't seem to have a really convenient way to just store blog links.