« Giugno 2005 | Main | Agosto 2005 »
27.07.05
Anyone want to buy a cello?
We're selling it in the UK (Oxford area) for 600 Sterling including bow and case. It's hungarian, full size and has a nice sound. Email susan.coleman at virgin dot net if interested - or leave us a comment.
See extended entry for pics
Posted by Giles at 15:24 | Comments (0)
25.07.05
We've got to get over paper
Paper - personally I hardly use it at all - it gets lost, it makes a massive mess, it isn't searchable etc... Mostly I use it only when forced to fill in pointless forms, for reading information that companies send me in the post, and reading books in bed.
Since the advent of email, the internet, the paperless office etc... you would have thought that printing and paper usage would have decreased but actually it has consistently and dramatically increased (see this and this).
So what are people on? I suspect it's a combination of
- Advertising by corporations with vested interests
- The fact that the techology has got a lot cheaper
- An attachment to tangibility - "if I can't hold it in my hand, it's not real".
About the lobbying and advertising, there's not a lot we can do, but as for the attachment to tangibility, I think we should try to get over it. I suspect that for most people a small change of attitude would be enough. It's a habit acquired in childhood to think that the real information is what's written on paper and what you see on the screen is somehow second rate and virtual. This really isn't necessary and once you get used to it, it's much easier to deal with things electronically. I admit that reading a book at bed-time is currently better with paper, but I'm talking more about the reams of paper we waste printing out documents at work, emails, photographs, newspapers etc.... etc...
Here's some suggestions
- Try to move correspondence into electronic form. If someone asks you to send a document by mail or fax, ask them if a scan sent by email would do. If someone gives you a printed document for comment, ask them to give it to you electronically. This takes the content off the paper circuit and puts further correspondence about the matter into electronic form.
- Don't write letters - it costs money and clutters the place up.
- Use commenting tools to exchange ideas and make suggestions live on the screen instead of writing things on a printout and then having to make changes to your docs a second time.
- Try to use online services like the inland revenue tax return system in the UK. No paper involved and you are guided through the irrelevant parts.
- use the google desktop search to find things electronically. It almost makes filing systems obselete as it's so quick to find virtually any document you've written among 10's of thousands.
Posted by Giles at 18:51 | Comments (0)
18.07.05
Creation science science
What motivates human beings to start from a preconceived and possibly meaningless pseudo-premise and devote their whole life to trying to get others to agree it's true? Let's get some of these freaks (check out chapter 2) into the lab and see what makes them tick. We might find out something interesting and extremely useful about religious extremism and how to foster open minded dialogue.
Even Bush is on the case now: "Both sides ought to be properly taught . . . so people can understand what the debate is about ... Part of education is to expose people to different schools of thought. . . . You're asking me whether or not people ought to be exposed to different ideas, and the answer is yes." This is a very insidious argument. Should we also teach people ptolemy's ideas on astronomy as a "school of thought"?
Posted by Giles at 14:52 | Comments (0)
11.07.05
When do you hear the fridge stop: time in consciousness
I've seen some very interesting stuff recently about the relativity of time in consciousness. It confirms some things I've long suspected.
I noticed that I often appear to start listening to the noise of the fridge humming just about 3 seconds before it stops. This of course isn't really possible, so what it points to is a time buffer in consciousness.
This is confirmed by recent experiments in neuroscience - here's a transcript from a Reith Lecture by VS Ramachandran.
"So we've talked about hysterical patients with hysterical paralysis. Now let's go back to normals and do a PET scan when you're voluntarily moving your finger using your free will. A second to three-fourths of a second prior to moving your finger, I get the EEG potential and it's called the Readiness Potential. It's as though the brain events are kicking in a second prior to your actual finger movement, even though your conscious intention of moving the finger coincides almost exactly with the wiggle of the finger. Why? Why is the mental sensation of willing the finger delayed by a second, coming a second after the brain events kick in as monitored by the EEG? What might the evolutionary rationale be?
The answer is, I think, that there is an inevitable neural delay before the signal arising in the brain cascades through the brain and the message arrives to wiggle you finger. There's going to be a delay because of neural processing - just like the satellite interviews on TV which you've all been watching. So natural selection has ensured that the subjective sensation of wiling is delayed deliberately to coincide not with the onset of the brain commands but with the actual execution of the command by your finger, so that you feel you're moving it.
And this in turn is telling you something important. It's telling you that the subjective sensations that accompany brain events must have an evolutionary purpose, for if it had no purpose and merely accompanied brain events - like so many philosophers believe (this is called epiphenomenalism) - in other words the subjective sensation of willing is like a shadow that moves with you as you walk but is not causal in making you move, if that's correct then why would evolution bother delaying the signal so that it coincides with your finger movement?
So you see the amazing paradox is that on the one hand the experiment shows that free will is illusory, right? It can't be causing the brain events because the events kick in a second earlier. But on the other hand it has to have some function because if it didn't have a function, why would evolution bother delaying it? But if it does have a function, what could it be other than moving the finger? So maybe our very notion of causation requires a radical revision here as happened in quantum physics."
The New Scientist recently carried an article which argued that our perception time is just one of many possibilities which happens to be evolutionarily advantageous. We could store up all events perceived and view them backwards, but this wouldn't help us for example in catching a ball... (or a fly if we are a frog).
I'm impressed with the British reaction to the bombings last week. There's no hysteria - calls for bombing innocent people in far-off countries, wall-to-wall press coverage, and not too much exaggeration. There was a call at my office among the british to have some minutes silence for the victims, but (not to diminish their suffering but...) the guy in charge just said - we didn't have a silence for the victims of the 2003 heatwave (30,000 of them) or the linate plane crash - so why this? - and we decided not to do it. I think the word *terror*ism is an important clue to how and why these things happen. With a relatively low cost action, a very small number of people cause practically a whole population to be terrified. In reality the chances of being hit by one of these guys in the average year is similar to winning a substantial amount of lottery money, dying in a road accident, and much less than a number of other risks kicking around. So well done to Blair for his life-goes-on stance.
A bit of Java which should have been on some mailing list somewhere but wasn't thereby causing me an afternoon of brainstrain - so here it is for googlers:
When you create a Swing JTree using the adapternode pattern - see here: Java Web Services Tuturial
This is a related problem but not the same:
Basically this is a method of displaying any XML or other document which is structured hierarchically in a very quick way by overriding the child, childcount etc... methods of the JTree node so that they get the children etc... according to the way the hierarchy is defined in your format. For example I used it to display an XML schema document which for DOM purposes is flat.
Anyway the problem is that it produces a kind of a stateless tree - in that each tree node only knows about its children and the whole tree doesn't seem to know about the paths it contains from root to leaf (although it does seem to know how many nodes it has in total). So it seems that it only works out the child node as and when you expand them. So the problem I had was how to expand and select a given path which I had matched in a list in the GUI.
This is the only way I could find to do it:
.....
//Create the treepath recursively using the recurseConstructPath method below and convert the vector to an array (otherwise you don't know how long it is before hand)
Vector treePathToSelectVector=recurseConstructPath(selectedListElement,new Vector());
TreePath treePathToSelect=new TreePath((TreeNode[])treePathToSelectVector.toArray(new TreeNode[treePathToSelectVector.size()]));
//Then recursively match and expand the visible rows as you go through the paths:
recurseExpandTree(treePathToSelect,treePathToSelect.getPathCount()-1,parent.parent.dataSchemaTree,0);
.....
Using these methods:
//Get the tree path recursively. Note that you have to add a parent() method to the Adapter pattern to make this possible - just pass the node itself to the child() method's adapternode constructer
private Vector recurseConstructPath(TreeDataTypeModel inputNode, Vector treePathVector){
treePathVector.add(inputNode);
if(inputNode.parent()!=null){
treePathVector=recurseConstructPath(inputNode.parent(),treePathVector);
}
return treePathVector;
}
//Go through all the currently expanded rows and if it matches that level of the path, expand it. Then do it again until you get no match. Note that the path nodes go backwards!
private void recurseExpandTree(TreePath path,int pathIndex,JTree tree,int startRow ){
int matchRow=0;
for(int i=startRow;i
tree.setSelectionRow(i);
tree.expandRow(i);
tree.scrollRowToVisible(i);
matchRow=i;
break;
}
}
if(pathIndex!=0){
recurseExpandTree(path,pathIndex-1,tree,matchRow);
};
}
Posted by Giles at 09:35 | Comments (0)
04.07.05
over 50% of marriages in the US end in divorce
This seems to be quite a simple fact and I've heard statistics like this many times. But when I think about it, it doesn't make a lot of sense. How do they know how many current marriages are going to end in divorce? If they're taking a sample of marriages which have already ended, then that's very misleading because all those people were married a rather long time ago. Are they perhaps taking a 10 year period to see who, of those who got married in that period, are already divorced. But then some of the people left over might also get divorced in the future.
Well - I did a bit of googling to find out what they actually mean by this. According to the US National Center for Health Statistics (part of the Centers for Disease Control and Prevention), in 2001 the per capita marriage rate was 7.8 marriages per 1,000 people (0.78%). (http://www.cdc.gov/nchs/fastats/divorce.htm) This means that for every 1,000 people living in the US, 7.8 marriages were performed during the year 2001, or 15.6 individuals got married. (http://www.divorcereform.org/rates.html) The divorce rate was 4.0 divorces per 1,000 people (0.40%), or 8.0 out of every 1,000 people got divorces during 2001. (http://www.cdc.gov/nchs/fastats/divorce.htm)
So what it really means is - the number of divorces per capita is over half the number of marriages per capita. That possibly makes the statistic worse than it seems at first sight because the number of divorces is based on a smaller population (assuming the US population is increasing and the number of marriages is not going down - which it does not seem to be - see http://www.divorcereform.org/98-00divorces.html). It also doesn't, as it appears to, give you likelihood for divorce for a current marriage, because the divorce rate now is for all people who got married in the last 65 years.
Posted by Giles at 22:40 | Comments (0)