For what its worth I've been writing an interpreter for
mathematical equations in Java. The principal problem here
is handeling different types of vectors and matricies
(i.e. 2,3 & 4 dimensional vectors, 2 by 2 matricies etc).
For each datatype I've used a different stack, so three
different stacks for the three different vector types,
and 9 stacks for matricies. Theres also corresponding
heeps for each data type which are allocated during
the compilation stage, eliminating the need for
any object creation during evaluation.
The benefits of making Java fully open source therefore seem overrated. Isn't the availablity of the source most important? Or perhaps I'm misunderstanding something...
For most users I'd agree with you, for me personally
the current way Java is licensed seems fine.
The one thing which could happen is that people could fork
the source to try some language extensions, something
like adding aspect oriented programming (aspectJ)
or adding in operator overloading. Sun could keep
their community process for the main branch.
A lot of this just seems to do with the name
Sun could feel good about being "open source"
GNU can feel good about another OS project, IBM
can feel good about making it happen and Microsoft
can feal confused. But it does not mean anything
really changes for developers.
Maybe there needs to be a distinction between
and Open Source License and an Open Source Development
Process. Just because the code is open source
does not mean they need to follow the same development
model as Linux.
Postition of chair, keyboard and monitor is very important.
By simply rasing my monitor by 6 inches greatly reduced
my RSI. With a low monitor you tend to hunch down creating bad posture
a high monitor means that your head is level and instantly
produces a better posture.
So much of the conversation just seems to be about compatability,
just trying to make OO as good as MSO. Why such low goals?
Surely the community should be going for better product,
products which stand above what MS can do, then people
will start switching.
We are starting to see good moves in this direction,
Mozilla is better than IE, Eclipse rocks. But lots
of the other stuff is still playing catch-up.
Theres lots of things which could easily be improved.
Get an intelegable help system (i've yet to
find anything useful in the MS help). Get some
good looking chart. Major fix needed for Excel
as its way behind the capabilities of the
serious numeric programs. Work on some better
DTP like features in Word (personal fav would
be a good way to print A5 booklets).
The open source movement does have a lot going for
it, lots more eyes, brains and ears. What are the
features which bug you, thats where to target.
I think a lot of it is historical. Back in the days of the first calculators it was easy to write a RPN calculator, in a very small number of bytes.
For the + operation its just
pop reg2 pop reg1 reg3 = add reg1+reg2 push reg3
The code required to parse brackets and corectly interprete
precedence of + and * is much bigger.
it is similar to the "music is math" argument. At the base level, yes it is. But there is just something extra (they "creative juices" and "unique style" you mention) that transcends music's mathematical base.
This reminds me of a conference called "The Art and Science of Baysian Image Analysis". Just because its maths does not mean that there is no space
for the creative juice. In Baysian image analysis (using techniques similar to those in modern spam filters to interpret images) there are many ways
to tackle the same problem. In it, as in many other mathematical/scientific fields, there is no one right solution, instead there is more than one
way to do it. Thats where the art comes in.
The one thing Iliked about objective C was that it didn't need a manual. The whole things could be described in a few web pages. No 700 page nutshell book just three or four pages. Now thats a neet language.
Another good book about infinity is "Infinity and The Mind"
by Rudy Rucker. This book explores the transfinite cardinals
in great depth, theres are other notions of infinity which are
"bigger" than the infinity of the integers or infinity of reals.
Its got some excelent stuff about the ultimate infinity,
i.e. the concept of infinity which is bigger than every
other concept of infinity. We can not actually give a precise
way of defining the ultimate infinity but there is some
truely head spinning stuff about some very large infinities.
Rudy's written several fiction books as well, "White light"
is a great romp through infinities examining Hilberts Hotel
and how to climb an infinite mountain in a finite time.
Java still lacks value classes and operator overloading, making it a poor choice for applications involving numerics or graphics.
I don't think this is a great loss, for serious numeric or
graphic code speed is the ultimate goal. A bit of verbose
language is a small price to pay. I haven't played too much
with generics in C++, but it did seem to involve a lot of
excess object creation which isn't too good for speed.
JEP - Java equation parser jep.sourceforge.net
just a simple parser for equations, but can be
used as the first steps for a CA program.
SingSurf - Draws singular algrbraic and other surface.
SingSurf.
Javaview - JavaView
a platform for 3D mathematical graphic.
It would be really nice to see a good open source
framework, which allows easy extension into domain
specific areas. I can't really see it happening.
Maybe the best thing is to work on interopability
of the packages, say through the OpenMath or MathML
systems.
A friend of mine has been doing bird surveys on this
very subject in the UK.
The problem can be reduced by intelegent placeing of
the turbines. Many species follow the same routes every
day, for instance the route from the nesting ground
to the feeding ground. If you put a turbine on such a route
then the chances of killing birds is higher.
Hence the need to accurately survey the bird population
around proposed wind farm sites and work out their
routes.
As I understand it, several proposed turbine sites have been
rejected as they are on the routes of certain endangered species.
Netbean it's also a system resource hog
Much as I like eclipse I have found it to gobble memory.
On my system 128 Meg Ram, if I run eclipse and mozilla
I'm basically out of ram and spend half my time listerning
to the disk thrashing.
I guess this has more to do with java than eclipse.
From the above link. MM03 = McIntyre and McKitrick, MBH98 = Mach et al.
The MBH98 reconstruction of Northern Hemisphere temperature since 1400 AD is an important result that certainly deserves scientific attention. MM03 obtain a rather different result to MBH98, which - if it were correct - would be very important.
Unfortunately neither MM03 nor the journal in which it was published took the necessary step of investigating (with Mann, Bradley or Hughes) whether the difference between MM03 results and MBH98 could be explained simply by some error or set of errors in MM03's use of the data or in their implementation of the MBH98 method. This should have been an essential step to take in a case such as this where the difference in results is so large and important. Especialy when the MM03 results, regarding a warm 15th century, were also at odds with the many other reconstructions that have been published, not just at odds with MBH98. Simple errors should first be ruled out prior to publication.
An important paper that re-examines historical
climate data was published on 28 October in the
respected journal Energy & Environment.
I must say I have my doubts about the "respected
journal Energy & Environment". Looking through the
list of abstracts they all seem to be of the greenhouse
sceptic camp. I've never heard of the journal before
it is published by a very little known publisher.
Certainly not on a par with Nature.
The editor is interesting on her webpage
she quotes
"Environment becomes fashionable and I turn sceptic".
Could I expect an unbiased paper in this journal.
Only one editor? Most journals I know have half a dozen if not more.
A lot of her own papers are published in her own journal,
just a little bit odd!
The first author is not attached to an institution, the second
is in an Economics department (not where I would expect to
find an experienced climate modeller). No funding was recieved for
the study.
The tone of the paper is strange, not what i typically
expect to see. It seems intent on rubishing the data
in every way possible. Most of the errors seem to be
minor, a one year slip in time (as data is agrigated over a year
it could be argued that this slip could just be a sampling error).
Some questions about missing details of methodology used
(if you've ever written papers with tight page limits
you'll know that some details always get left out).
The main results fig 7 and 8 are interesting.
Their results show broard agreement with
the critisied paper for years 1575-1980. It is only
the years pre 1575 where there is marked disagrement.
Its no suprise that most errors appear in earlier
years when the data is presumably less relaible.
I must say I'm sceptical about their pre 1600 data
that is one heck of a temprature drop, do other sources
show similar results?
In particular the paper backs up the Mann's results
from 1600 onwards. Especially the last 100 years where
both show a rapid increase in temprature.
Amounting to.3 to.5 degree change in global temprature
in the past 100 years.
So is this another slashdot classic from the bottom of the barrel?
Let me know when it appears in nature.
At my old uni they had a very similar project doing
3D laser scans of building and meshing them with
the visible pictures. Have a look at
Resolve Project.
I'm working with a group of people trying to
put a colaborative plant database together.
Draft Version.
The idea is to put together a large dataset of plants
together.
Wiki's seem good, but they miss one important aspect,
structure to the documents. Details about plants neetly fall
in to a number of catagories
Latin/Botanical name, Common name, growing habit, etc.
What I'd like to do is take wiki type concept but add more
structure to the data. This could help with searching.
Also some fields such as height have numeric values
and it would be great to search for plants with a specific height.
Anyone come across such ideas or software which could do such a thing?
BTW I'm suprised how down most slashdotters are on
colaborative documents. There are some really good colaborative encyclopedia
around wikipediaPlanet Math.
So whats wrong with OpenContent!
I think we need to make a clear distinction
between Open Standard and Open Source. They
are very different beast and should not always be
confused. The Report seemed to be more enthuisatic
about the standards than the source.
Open Standards are all about interoperability
creating a level playing field where companies
can compeat to produce the best readers and
writers of the standards. Consumers and
Govs are free to choose which suits them best.
This is one of the reasons the web took off
as html was essentially an open standard, even
though there were no open source browsers
about in the early days.
Open Source is a different beast.
I don't think the the benifits for a company
to open source its products are as clear.
Yes there are advantages with transpanancy for
govemental use. Yes its great for hobbyiest,
probably great for products aimed at developers.
But the economic model is dificult, the viral
licencing can cause problems.
In general I'm much more passionatle
about Open Standard than Open Source.
Because, after all, taking the root servers away from bright, educated comp-sci longbeards who have nothing better to do than to make them run well, and putting them in the hands of MBA bean-counters who don't know what TCP/IP is, is a sure-fire way to improve reliability.
How are we ever going to have well ballanced discussions on slashdot when the
editors come out with such dicdely biased takes on the subject.
Data labelling and ACID complience seem quite good
on
CNet on WinFS
·
· Score: 1
Well everyone else seems to think this is a bad thing
but I can seem some postitive aspects.
Using SQL might give ACID complience which could
protect data. No more lost/corrupted files when Word
crashes.
Data labeling through xml seems like to could offer
some advantage. I've often wanted to add a brief note
to my files explaining what the file is. Say for a jpeg
image have a text description of the image. May be I could
add some tags or Key Words to the meta information
about the file. This would help me search.
Whilst a tree structure for files/directories is probably
the easiest way to organise things it does have
limitations. Symbolic links don't always do the job.
Find is currently way too slow maybe something with
the simplicity of google's search box is the easiest
way to get to your data?
Maybe the fs will be able to cope with different namespaces
providing a simple interface to all different resources.
It seems like MS are trying to do something
inovative and this for one should be aplauded.
Why not go down the mySQL dual licence route?
Cisco quite rightly want to keep there own
stuff closed, thats their free right to do so.
But of course they should not get if off
the back of others. So whats wrong with
just giving them the freedom to use stuff
how they want as long as theres appropriate
payback to the developers. An appropriate
licence fee paid to some nuteral body could
be any easy resolution to these problems.
For myself as a developer of small standalone
libraries, I'd like to make my software free
so others can build on it but I also want to
let comercial orginizations use the software as
well. Without a dual licence no closed buisness
will be able to use it. Leaving them without a
product and me broke.
I'd like to draw a line at the API for the
library, if the modification is on the
library side of the API then thats LGPL
if its on the other side thats your buisness
it is not my right to dictate how they should run
their buisness (at a rpice of course).
If something is going to be free to use
then it has to be free of imposing an idiology.
Physisits have been doing something similar for ages.
Have a look at http://arxiv.org/.
Most phyisics papers appear here first, only later going
on to paper publishers. The big bifference between the two
is arXiv has no reviewing process (its for pre-prints).
This does make things quicker
which seems to be what physist want, but might have impact on
quility of papers.
When an "out of the office" auto-reply comes back on one e-mail message, Colbert says: "Oh, we love those. They confirm that the address is active."
This should put to rest any remaining doubts about whether or not "unsubscribing" from spam lists actually works.
Maybe what is needed is a trusted third party, someone I could
send my remove me request to and actually beleive that something
would be done about it.
From my experience of running a high content website
(7000 pages about plants Plants For A Future
) is that fame is great for the first few years, but
it begins to ware thin as time goes on. I'm increasingly
looking for ways to try and gain some income for our
project from the website. I think this is close to what
Scott has to say The effect of fame on the Web is often to rob the author of his or her fortune through excess bandwidth charges.
I've not really tried hard enough yet but thinking of
BitPass, Google Adds (finally an add system which meets
my required level of intrusivness), donations (some hope!)
and some higher price downloadable content (a very small
income stream).
I must agree with Scott BitPass does seem to have a good
system.
Final point to ponder. Would I have bought Scott's comic
if I wasn't actually interested in how BitPass could work for me?
In a world of info overload would I pay some something
which was not vital for me?
Slashdot Mirror
For what its worth I've been writing an interpreter for mathematical equations in Java. The principal problem here is handeling different types of vectors and matricies (i.e. 2,3 & 4 dimensional vectors, 2 by 2 matricies etc). For each datatype I've used a different stack, so three different stacks for the three different vector types, and 9 stacks for matricies. Theres also corresponding heeps for each data type which are allocated during the compilation stage, eliminating the need for any object creation during evaluation.
A lot of this just seems to do with the name Sun could feel good about being "open source" GNU can feel good about another OS project, IBM can feel good about making it happen and Microsoft can feal confused. But it does not mean anything really changes for developers.
Maybe there needs to be a distinction between and Open Source License and an Open Source Development Process. Just because the code is open source does not mean they need to follow the same development model as Linux.
Postition of chair, keyboard and monitor is very important. By simply rasing my monitor by 6 inches greatly reduced my RSI. With a low monitor you tend to hunch down creating bad posture a high monitor means that your head is level and instantly produces a better posture.
We are starting to see good moves in this direction, Mozilla is better than IE, Eclipse rocks. But lots of the other stuff is still playing catch-up.
Theres lots of things which could easily be improved. Get an intelegable help system (i've yet to find anything useful in the MS help). Get some good looking chart. Major fix needed for Excel as its way behind the capabilities of the serious numeric programs. Work on some better DTP like features in Word (personal fav would be a good way to print A5 booklets).
The open source movement does have a lot going for it, lots more eyes, brains and ears. What are the features which bug you, thats where to target.
This reminds me of a conference called "The Art and Science of Baysian Image Analysis". Just because its maths does not mean that there is no space for the creative juice. In Baysian image analysis (using techniques similar to those in modern spam filters to interpret images) there are many ways to tackle the same problem. In it, as in many other mathematical/scientific fields, there is no one right solution, instead there is more than one way to do it. Thats where the art comes in.
But it would probably never get through a spam filter!
The one thing Iliked about objective C was that it didn't need a manual. The whole things could be described in a few web pages. No 700 page nutshell book just three or four pages. Now thats a neet language.
Rudy's written several fiction books as well, "White light" is a great romp through infinities examining Hilberts Hotel and how to climb an infinite mountain in a finite time.
Java still lacks value classes and operator overloading, making it a poor choice for applications involving numerics or graphics. I don't think this is a great loss, for serious numeric or graphic code speed is the ultimate goal. A bit of verbose language is a small price to pay. I haven't played too much with generics in C++, but it did seem to involve a lot of excess object creation which isn't too good for speed.
JEP - Java equation parser jep.sourceforge.net just a simple parser for equations, but can be used as the first steps for a CA program.
SingSurf - Draws singular algrbraic and other surface. SingSurf.
Javaview - JavaView a platform for 3D mathematical graphic.
It would be really nice to see a good open source framework, which allows easy extension into domain specific areas. I can't really see it happening. Maybe the best thing is to work on interopability of the packages, say through the OpenMath or MathML systems.
The problem can be reduced by intelegent placeing of the turbines. Many species follow the same routes every day, for instance the route from the nesting ground to the feeding ground. If you put a turbine on such a route then the chances of killing birds is higher. Hence the need to accurately survey the bird population around proposed wind farm sites and work out their routes.
As I understand it, several proposed turbine sites have been rejected as they are on the routes of certain endangered species.
Maybe there are leasons to be learnt about persuing IP claims, you may win the legal battle but loose the war!
I guess this has more to do with java than eclipse.
The MBH98 reconstruction of Northern Hemisphere temperature since 1400 AD is an important result that certainly deserves scientific attention. MM03 obtain a rather different result to MBH98, which - if it were correct - would be very important.
Unfortunately neither MM03 nor the journal in which it was published took the necessary step of investigating (with Mann, Bradley or Hughes) whether the difference between MM03 results and MBH98 could be explained simply by some error or set of errors in MM03's use of the data or in their implementation of the MBH98 method. This should have been an essential step to take in a case such as this where the difference in results is so large and important. Especialy when the MM03 results, regarding a warm 15th century, were also at odds with the many other reconstructions that have been published, not just at odds with MBH98. Simple errors should first be ruled out prior to publication.
I must say I have my doubts about the "respected journal Energy & Environment". Looking through the list of abstracts they all seem to be of the greenhouse sceptic camp. I've never heard of the journal before it is published by a very little known publisher. Certainly not on a par with Nature.
The editor is interesting on her webpage she quotes "Environment becomes fashionable and I turn sceptic". Could I expect an unbiased paper in this journal. Only one editor? Most journals I know have half a dozen if not more. A lot of her own papers are published in her own journal, just a little bit odd!
The first author is not attached to an institution, the second is in an Economics department (not where I would expect to find an experienced climate modeller). No funding was recieved for the study.
The tone of the paper is strange, not what i typically expect to see. It seems intent on rubishing the data in every way possible. Most of the errors seem to be minor, a one year slip in time (as data is agrigated over a year it could be argued that this slip could just be a sampling error). Some questions about missing details of methodology used (if you've ever written papers with tight page limits you'll know that some details always get left out).
The main results fig 7 and 8 are interesting. Their results show broard agreement with the critisied paper for years 1575-1980. It is only the years pre 1575 where there is marked disagrement. Its no suprise that most errors appear in earlier years when the data is presumably less relaible. I must say I'm sceptical about their pre 1600 data that is one heck of a temprature drop, do other sources show similar results?
In particular the paper backs up the Mann's results from 1600 onwards. Especially the last 100 years where both show a rapid increase in temprature. Amounting to .3 to .5 degree change in global temprature
in the past 100 years.
So is this another slashdot classic from the bottom of the barrel? Let me know when it appears in nature.
At my old uni they had a very similar project doing 3D laser scans of building and meshing them with the visible pictures. Have a look at Resolve Project.
Wiki's seem good, but they miss one important aspect, structure to the documents. Details about plants neetly fall in to a number of catagories Latin/Botanical name, Common name, growing habit, etc. What I'd like to do is take wiki type concept but add more structure to the data. This could help with searching. Also some fields such as height have numeric values and it would be great to search for plants with a specific height.
Anyone come across such ideas or software which could do such a thing?
BTW I'm suprised how down most slashdotters are on colaborative documents. There are some really good colaborative encyclopedia around wikipedia Planet Math. So whats wrong with OpenContent!
Open Standards are all about interoperability creating a level playing field where companies can compeat to produce the best readers and writers of the standards. Consumers and Govs are free to choose which suits them best. This is one of the reasons the web took off as html was essentially an open standard, even though there were no open source browsers about in the early days.
Open Source is a different beast. I don't think the the benifits for a company to open source its products are as clear. Yes there are advantages with transpanancy for govemental use. Yes its great for hobbyiest, probably great for products aimed at developers. But the economic model is dificult, the viral licencing can cause problems.
In general I'm much more passionatle about Open Standard than Open Source.
How are we ever going to have well ballanced discussions on slashdot when the editors come out with such dicdely biased takes on the subject.
Using SQL might give ACID complience which could protect data. No more lost/corrupted files when Word crashes.
Data labeling through xml seems like to could offer some advantage. I've often wanted to add a brief note to my files explaining what the file is. Say for a jpeg image have a text description of the image. May be I could add some tags or Key Words to the meta information about the file. This would help me search.
Whilst a tree structure for files/directories is probably the easiest way to organise things it does have limitations. Symbolic links don't always do the job. Find is currently way too slow maybe something with the simplicity of google's search box is the easiest way to get to your data?
Maybe the fs will be able to cope with different namespaces providing a simple interface to all different resources.
It seems like MS are trying to do something inovative and this for one should be aplauded.
For myself as a developer of small standalone libraries, I'd like to make my software free so others can build on it but I also want to let comercial orginizations use the software as well. Without a dual licence no closed buisness will be able to use it. Leaving them without a product and me broke.
I'd like to draw a line at the API for the library, if the modification is on the library side of the API then thats LGPL if its on the other side thats your buisness it is not my right to dictate how they should run their buisness (at a rpice of course).
If something is going to be free to use then it has to be free of imposing an idiology.
Physisits have been doing something similar for ages. Have a look at http://arxiv.org/. Most phyisics papers appear here first, only later going on to paper publishers. The big bifference between the two is arXiv has no reviewing process (its for pre-prints). This does make things quicker which seems to be what physist want, but might have impact on quility of papers.
When an "out of the office" auto-reply comes back on one e-mail message, Colbert says: "Oh, we love those. They confirm that the address is active."
This should put to rest any remaining doubts about whether or not "unsubscribing" from spam lists actually works.
Maybe what is needed is a trusted third party, someone I could send my remove me request to and actually beleive that something would be done about it.
I've not really tried hard enough yet but thinking of BitPass, Google Adds (finally an add system which meets my required level of intrusivness), donations (some hope!) and some higher price downloadable content (a very small income stream).
I must agree with Scott BitPass does seem to have a good system.
Final point to ponder. Would I have bought Scott's comic if I wasn't actually interested in how BitPass could work for me? In a world of info overload would I pay some something which was not vital for me?