Powered by glolg
Display Preferences Most Recent Entries Chatterbox Blog Links Site Statistics Category Tags About Me, Myself and Gilbert XML RSS Feed
Monday, June 10, 2019 - 19:14 SGT
Posted By: Gilbert

Mixed Bag

"In a country as rich as Singapore with no absolute poverty, economic inequality just means some people have become richer than others..."

- Brook & Watkins, for The State's Times


Gee, I suppose the stubborn refusal to define a poverty line might have helped justify the bold assertion on a total lack of absolute poverty from the Ayn Rand Institute here. The libertarians did make their point about how inequality might not be undesirable as long as everyone gets by, but the ease with which that last assumption got accepted was bemusing nonetheless. This did at least get questioned in the forums, which pointed out Prof Tommy Koh's indirect acknowledgement of absolute poverty in last year's IPS conference - coincidentally, the one in which he skewered The State's Times for biased reporting.

Tellingly, the above article came right as a couple of local Catholic churches announced that they were allowing homeless to bunk in for the night - which could be interpreted both ways in the existence of absolute poverty debate. The move towards mixing residents of varying economic statuses in the same block is probably a step in the right direction - and as it turns out, it doesn't even cost much if anything to zhng one's HDB flat to condo standards. Automated bicycle storage may however be an example of public sector funding inefficiencies - how many are there willing to cough up S$48 a month for vaulting their bike? At that pricing, they could get a whole new bike every few months, and it's not like they get the convenience of islandwide availability either, as was the case with the mostly-exited rental services.

As it turns out, trying to help the needy out can be a thankless task here, from how an undergrad's initiative to collect five-cent coins got roundly panned. But, as the government's selling: does a problem even officially exist, if one simply chooses not to own up to it?


Money Management

After receiving a call from a DBS representative touting their Multiplier account some weeks back, I agreed to meet up with him to go over the terms & conditions. The lure was tantalizing - rather than a paltry 0.05% p.a. interest on a regular savings account, the Multiplier account offers up to 3.80%! Given that I had been familiar with poor fixed deposit rates of 2% or less from a decade or so ago, I could not help but be curious about the catch here. As it turns out, there are a few.

Firstly, the interest rate applies only for the first S$50000 (recently extended to S$100000, with some further conditions). Secondly, the maximum 3.80% interest rate applies only if over S$30000 in transactions are made per month, with four conditions being met: salary crediting (compulsory), and three of credit card spending, mortgage payments, insurance and investments. If the transaction volume is lower, or if fewer than three of the optional conditions are met, the interest applied decreases accordingly.

From this, the bank's intentions might be deduced: the generous interest offered is a loss leader for their ancillary services, in particular insurance & investments. Consider the analogy: if one spends on something he doesn't really need, just because of the generous cashback on his credit card, does he really benefit? Note that the maximum interest obtainable per annum - S$3650 from an effective 3.65% interest rate - can realistically only be obtained by taking out some mammoth policies; from another direction, how many people chasing interest on S$100k, naturally spend over S$30k a month?

That said, Singaporeans are, if nothing else, always up for maximizing offers; the relevant EDMW thread has tons of tips on how to best exploit the system, which extends to cycling cheap term insurance policies yearly to satisfy the insurance component. Others seem to have been suckered into dabbling with equities for the investments component, at which I can only imagine some smug moustache-twirling from the bankers. Like, the additional bonus interest from the second S$50000 for a third category is but S$150 a year; it may not be nothing, but does optimization to that extent make sense?

Well, it wasn't as if the representative was much interested in selling me term insurance at any rate, as it quickly became apparent that he was angling more a more lucrative whole life-type signup, same as the rest of them. In any case, the unsuccessful spiel concluded with the revelation that an in-person meeting was not, after all, necessary for account activation - that could be achieved via the i-banking app. Fair deal for an extra thousand or so a year, I'd say.


Virtualization Marches On

In another blow to the physical (e.g. land, retail space, electronic hardware) as commodity, Google has announced its Stadia cloud gaming-as-a-service. Before this, gaming had arguably driven much of the computer hardware market, most significantly with graphic processing unit (GPU) cards, but also other components such as RAM and the CPU - the everyday user running Word and Excel would probably do just fine with a PC from fifteen years ago, really. And this is before we get to the pricey dedicated consoles: the Xboxes, PlayStations, Nintendoes etc, that get obsoleted every so often.

Enter Google and the age of abundant fibre bandwidth. Stadia's approach is for Google to deal with all the hardware headaches invisibly - no more agonizing about whether to upgrade GPUs or consoles. Instead, all users have to do is to connect to Stadia via Chrome (they've implemented Remote Desktop directly as a Chrome tab recently, which I'd suspect is related), and select a game (which also eliminates installation, compatibility and disk space headaches). Any user instructions (e.g. keystrokes, mouse clicks) are then sent to the Stadia servers for interpretation (greatly reducing hacking opportunities), with the full HD screen output streamed live back to the user, kind of like with Netflix.

Let's view this development in light of the last big leap in games distribution: Valve's Steam. Launched back in 2003, Steam was itself driven by bandwidth availability, after they realized that having customers download games, instead of buying them on a CD-ROM, was feasible. The benefits were clear: no more worrying about losing physical CDs, and far less bother with bug fixes and patch applications. Unsurprisingly, Steam's rise and dominance has led to the impending demise of brick-and-mortar games retailers such as Gamestop, who seem destined to go the way of Blockbuster.

Retail hardware such as GPUs and consoles remained afloat, though, since Steam merely distributed the software, and required the processing to be done locally. This is no longer the case with Stadia, and if their cloud gaming paradigm takes off, one would expect the end of the retail hardware market, since all users would require would be input devices such as a keyboard/mouse/gamepad, and screen.

Personally, I wouldn't have guessed that bandwidth had progressed to this extent, and I'd wager that purists might bemoan the added latency on twitch reaction games (the sort that checks out refresh rates on monitors, at least). The broader issue here would be the increasing trend away from local control; yes, if there's any firm that can sponsor hardware for everyone - Stadia will have a free tier - it's probably Google (see: Gmail). However, this would appear be leading to a world where an individual's data and virtual identity is entirely under the purview of corporate interests. And from how Google has been handicapping ad blockers to safeguard their main revenue stream, I'm uncertain if they can be trusted on this trend.

One thing I'm willing to bet on, though - watch for China to release a Qiuchang clone in a year or two.


Stemcells vs. Organs

The latest outrage for the American set is on the TRUMP administration's curtailing of funding for medical research that utilizes tissue from aborted fetuses, which ties neatly into the whole abortion uproar that's been making the rounds (note: a former Hwa Chong alumni is all for empowerment in this regard). The reaction was expectedly near-uniformly negative amongst the usual Reddit bunch. I reproduce a selection of representative comments here:

"Polio vaccine... was created decades ago via fetal tissue research. Experts estimate that tens of millions - hundreds of millions of lives have been saved due to said vaccines."

"So instead of it being used in research that could save lives, it's going to be discarded. And... this is a victory to pro lifers? I'm honestly perplexed."

"Drumpf's crusade against science is going to cost lives!"

The most pertinent parallel that occurred to me did not, however, warrant a mention - that of organ harvesting from executed prisoners in China; the abhorrence against this is so strong amongst scientists, it seems, that there has been a call for the retraction of over 400 scientific papers over concerns that such organs had been unethically used for research.

The insistence by Chinese authorities that the prisoners were legally condemned to die in any case, and that their organs might as well be put to life-saving use, does not seem to have found nearly as much purchase amongst Redditors; an obvious objection would be that if organ harvesting were officially legitimized - and profited from - then the relevant authorities might just employ the death penalty more freely, to keep up with demand. Indeed, this profit-seeking motive has been observed with the private prison industry in America.

Given this, is it not out of the realm of possibility that a demand for fetal tissue could nudge some medical institutions towards encouraging abortions, more than they would otherwise have had? Whither the distinction between unliving tissue, and unliving organs?


Grants & Tenure

Some interesting recent research from the MIT Sloan School of Management on how academic grants are awarded suggests that "broad" and general terms are positively correlated with better outcomes, as compared to "narrow" topic-specific words. Examples given were "bacteria"/"detection" (broad) as compared to "community"/"health" (narrow). Frankly, I'd have guessed it went the other way around; anyhow, the message of the paper was on gender: female scientists, so it seems, tend to use narrower terms than males, which resulted in lower grant proposal scores. This was deemed a discovery of sufficient impact to have been featured in Nature and Science.

While I have admittedly been unable to access the paywalled working paper itself, my own suspicion is that the division may be less broad/narrow, and more technical/non-technical; I'll be keeping an eye out for the final published version. That said, academia was always supposed to be more a marathon than a sprint - former U.S. President Jimmy Carter has just made tenure at Emory after 37 years there, at the tender age of 94. What's the big hurry, huh?



comments (0) - email - share - print - direct link
trackbacks (0) - trackback url


Next: Cab Chat


Related Posts:
Whiter Than White
Light It Up
Open Bookame
A World In One
Softly Does It

Back to top




Copyright © 2006-2019 GLYS. All Rights Reserved.