I've been skimming a giant book of software components. (I read it a long time ago.)
I remember telling a friend about components, & he asked ``what do they buy you'', or maybe he pointed out that you can do without components all the things you can do with them. I wasn't foolish enough to disagree that non-component software could accomplish everything that components do except maybe some ``essense of components'' subtleties (& vice versa, if we are considering subtleties), but I still said components did some things different or better.
But now I recant. I'm currently undecided between these two ideas:
Like shared libraries, components could help you save a lot of disk space. Okay, that's an advantage.
The benefit of binary distribution by itself is only when your target platform doesn't have a compiler. That's the reality with Winders, but it doesn't have to be. Every Winders computer could come with a compiler. If companies were worried about revealing their proprietary source code, Winders (& other computers) could offer a standard intermediate language & a compiler for it. The software developers would compile to the intermediate language & ship that. (In fact, this is one of the unrealised features of dot-Net.7.1 Also, there are many advantages to compiling on the target machine. )
The other benefit of shared libraries & components is that you can change the behaviour of your program without recompiling. If ``change the behaviour'' means ``fix bugs or add features'' (as opposed to ``entirely change the focus of the program''), I'm not sure it's all that useful in a production environment. Sure, it is definitely cool, but is it necessary/practical? I'm not sure.7.2
So I'm undecided between these two points of view.
Overall, I now dislike the concept of components because the only reason programmers think they are the best thing since slice bread (& before web services) is hype &/or the programmer's own misunderstanding of the pure & practical concepts of ``what needs to be done'' and ``how do get it done simply''.
Component systems are still nifty from a not-necessarily-practical ``how does it work'' point of view.
Yesterday, I read about the United States's ``War on Terror'' to refresh my memory of the facts. I tried to keep notes so it would require less work to refresh my memory the next time. I also wanted to find important facts & document them with refernces so I could argue...er, I mean discuss the issues with people who can't see the light...er, I mean, people who don't share my opinion.
Okay, I'll be serious now.
I learned that making notes of such a complex issue is really difficult. There are so many people & organizations involved, organizations containing other organizations7.3, disputed alliances, events with multiple effects, & a wide range of time resolutions.7.4 More importantly, sometimes you want to see all events that happend during a period of time (such as on 2001 September 11), while other times, you want to see events involving a particular person (such as Colin Powel). My usual technique of taking notes isn't sufficient.
I'm not saying the issue is too complicated for a human mind to comprehend. I'm saying that it's too complicated for a human mind to comprehend when that human mind has a day job. More importantly, it's so complicated that a human mind with a day job, after finding some important conclusion, will not have enough time to convince another human mind (also with a day job).
This applies to issues other than this ``War on Terror''.7.5 It could apply to socialized health care, the economics of retirement, and convoluted & questional campaign funding7.6 to name just a few issues that are current in the politics of the USA.
Maybe the way to understand complicated issues like this one is to track events, people, & organizations on a finer scale & use software to view those things in different ways.
People & organizations might might be connected with links such as ``loyal to'', ``held responsible by'', or ``hates''.
Events might be connected to the involved people & organizations. They could also have a time, & the resolution of the time would be significant. For example, 2005 July would mean ``some time during the month of 2005 July''; 2005 July 4 would mean ``some time during the day of 2005 July 4''; and ``2005 July 4 21:01'' would mean ``some time during the minute of 2005 July 4 21:01''. I suspect resolution down to the minute would be sufficient for all political issues.
The way I've described it, with ``links'', it sounds like I'm thinking of a relational or networked database, but I'm sure a quick & dirty (& sufficient) implementation could use plain text if you were consistent in how you spell names. For example, George Washington would always need to be ``George Washington'', never just ``Washington''.
If each event was a single line in a file, starting with the date in ISO format (such as ``20050704T2101'' for 2005 July 4 21:01), you could use standard un*x tools for a decent implementation. Here are some examples:
|grep -i "bin laden"
|sort''. This would include everything Bush said about Bin Laden, but also any event involving both men, including things Bin Laden said about Bush or the time at the party when Bin Laden & Bush shared a bong.7.7
You might use grep & sort like this to generate a report that contained all events in one section & then a section for each person or organization involved that showed the events involving that person or organization.
The text of the events could be HTML, complete with links to the web pages that discuss each event.
Well, hell. When I started writing this, it was to show how difficult the project would be & leave my thoughts on how to do it if I ever had the time. Now it looks simple.
The facts appear to be:
As of 2005-Jul-07T7:50, the only claim of responsibility comes from the web site of ``The Secret Organization of al-Qaeda in Europe'' (SOAQE)7.9. SOAQE says the attacks were in retaliation for Britain's involvement in Iraq & Afghanistan. (Note that this is in contrast to Tony Blair's explanation for the attacks.)
On 2004 March 11, there were similar attacks on the trains in Madrid. They killed 191.
Senator Sam Brownbag(?), Republican from Kansas, on the radio at this moment claims that the attacks were because the G-8 summit was trying to think of ways to help poor countries in Africa. He said the poor countries in Africa will be an important tool to combating terrorism. I do not agree with his claim. Blair's explanation & the claim on the SOAQEd web site make more sense.
Spanish officials note the similarity between today's attacks in London & last year's attacks in Madrid. They say Madrid is now on maximum alert. They are closing the hen coop more than a year after the fox has visited.
Cooper & Niven's short story was excellent.
What does it mean to ``secretly monitor'' a broadcast?
I guess I can understand why a Republican would keep tabs on the Diane Rehm Show, since she interviews some people who sometimes criticize our government mildly, but Tavist Smiley? What about that show could possibly irritate anyone who wasn't a KKK member? I mean, really.
I saw the movie the other day. Here are some thoughts about it.
Bill, the eye patient & main character, begins the movie as does the main character in 28 Days Later. The viewer has a bigger clue about what's happening in Triffids than in 28 Days, but the stories of the two main characters begin similarly. TheywWake up in an empty hospital & don't know where everyone went.
Bill's story begins with irony. When we first see him, he is effectively blind due to a head bandage. The next day, when almost everyone is blind, his sight is fine.
Like modern zombie movies7.10, Triffids asks what would happen if a lot of people failed to perform their roles in society. In Triffids, those people become mildly detrimental because they bumble around & might cause a sort of small, bumbling riot. In zombie movies, people fail to fill their roles, and they become actively hostile.
The plant attack scenes remind me of the Doctor Who story, Seeds of Doom. In Triffids, they even use the same dark green colour on the skin of some of the victims.
So the triffid plants arrived during an earlier meteor shower? They were harmless until the meteor shower which we see in the movie? Were these meteor showers an attack? Or just very bad luck?
Killer plants arrive from outer space. Reminds me of Invasion of the Body Snatchers.
There are three possible results if so many people went blind:
I am genuinely curious about what happened to the world of Triffids after the movie ended & while that world learned to cope with all those blind citizens.
I notice that characters in both these animes refer to robots as ``boomers''. Interesting coincidence.
Oops. Not such a coincidence. Chiaki Konaka wrote the screenplay for Bubblegum & has the full writing credits for Parasaito.
I wonder how the original Bubblegum Crisis referred to robots. It's been so long since I saw it that I don't remember if it referred to them at all.
Compare & contrast ``boomer'' & similar terms:
|android, realien||Tomorrow's Eve||novel||1880?||Mathias Villiers de l'Isle-Adam|
|robot||Rossum's Universal Robots||play||1920||Karel Capek|
|droid||Star Wars||movie||1977||George Lucas|
|replicant||Blade Runner||movie||1982||Philip K. Dick|
|boomer||BC: Tokyo 2040||TV show||1998||Hiroki Hayashi|
|boomer||Parasite Dolls||OVA?||2002||Nakazawa and Yoshinaga|
Mark Whitby, a commuter, told BBC television that he was sitting on the subway train reading the newspaper when the fugitive and the police burst through the open doors. The man was a stocky young Asian, possibly of Pakistani descent, and wore a baseball cap and a padded winter-style coat, Whitby said. After a melee in which the suspect fell or was wrestled to the floor, an officer opened fire at close range as passengers screamed and cowered, Whitby said.
``I heard a load of noise...people saying, 'Get out, get out','' Whitby said. ``I saw an Asian guy. He ran on to the train, he was hotly pursued by three plainclothes officers...he half tripped and was half pushed to the floor and the policeman nearest to me had the black automatic pistol in his left hand. He held it down to the guy and unloaded five shots into him.''
At last someone has given me a good reason to use JOIN - portability. I disagree with the readability claim, & the dozens of programmers who tell me it's more efficient at run-time or that (get this) ``you just can't join two tables without JOIN; it simply won't work'' are foolish. But portability, that's a good reason.
Sometimes, I think that even nearly-pacifist me could indeed be motivated to assassinate...
I've been wishing we'd ditch Daylight Stupid Time & even time zones ever since I wrote the time-handling code for a telephony switch testing system in 1997.
In 2000, I had to write a similar chunk of code for a 9-11 system with a Russian programmer. As he realized the shittiness of DST & time zones, one day he shouted (in his thick Russki accent) ``Cursed time!'' (Hee hee. It still cracks me up.)
For the nitty-gritty of just how amazingly horridly unbelievably contemptibly distgustingly shitty time-related code must be due to a shitty time-keeping system, look at P.J.Plauger's implementation of C's ``time.h'' functions in The Standard C Library.
In case you're interested, there is a readable history of Daylight Stupid Time a http://www.timechange.com/dls/. It all goes back to some English gentleman who was personally insulted by late risers. So it's an example of the practical harm that can be done by a large ego.
I apologize for sounding like a know-it-all. I really am
^H^H^H^H^H^H^H^Hinterested in this issue.
Okay, I'm confused. The article sez: ``farmers complained that a two-month extension could adversely affect livestock''.
In what way would DST affect livestock at all? Shouldn't the farmer get up when the sun comes up (or something like that)? Why does the farmer, or his livestock, give fuck-all what the clock says?
I could understand (not necessarily agree with) complaints from Wall Street, politicians, restaurants, television networks, & others, but from farmers on behalf of livestock? Seriously, I'm confused.
Gene Michael Stover 2008-04-19