Older blog entries for Chicago (starting at number 23)

Hokay, longtime no post reason being that ive been working on various things to post about.
A query: Does anyone know anything about a radio system which transmits through the ground? I believe it was used by la resistance during the war, but I do not know what it is called or anythng... any ideas?

Interesting Problem 1
The search for a solution has taken me far and wide - from the depths of genetics to RTTY systems develioped as far back as 1832. Now I suffer from being unable to combine all this knowledge into something meaningful. Im struggling to get hold of relevent documents for what I want to do. Currently the best people are the RSGB, and I have quite a large collection of those books now. Unfortunatly if I wanted to build any of the circuits i would require a thousand and one valves, all of which where easily availible back in 1938 when most of these book where originally published.

Matt
Returning to the problem of no maze, but also no processor board, I have been getting the parts together that I will need to build the mouse. This should enable me to build the mouse soonish.

Orgy
I am currently trying to work out if it is worthwhilst continuing this project. Something makes me think iDunno will have other ideas about that, it being his idea and all. Perhaps with moving back to Norwich soon I will have more enthusiasm about the project.

How to fai... ahem pass your alevel computing project
This small document I have ended up writing for kaz. Ive been considering making it into a book and selling it for 35p (price of a marsbar). That would keep me in mars bars for a couple of years. (am I allowed to discuss about... comercialisms on advogato? probably not)

Life
Dont talk to me about Life.

argh!! no!!! im soo sorry - i cant even read my own diagrams right. ishamael and ladypine both have it right, and I originally did mean to make b = 111, only failed to read my own diagram correctly, so screwed it all up. Im now going back to corrct all my maths mistakes based upon that for the past week or so :(

Lesson learnt? I am a wuckfit at times

thanks to ralph, guerby and mbrubeck for pointing me in the right direction. (or at least a better direction). Ive currently given up on the Graph Theory aproach to finding a generic solution to the problem, as graph theory dosnt seem to be very ... advanced.. just a collection of theories which they seem to be having extreme difficulty prooving. Most of it seems to be just plain common sense, and the other half brute forcing. Indeed, the word "discovered" seems to be used a lot.

Anyways, I have actually come across hamming before - but I was always told that this was because of HAM radios .. ham being an .. how to say it? affectionate? term for ameateur radio member. These people where (or so i was told) the original pioneers of long distance networks via the radio, and of course that has a lot of intereferance so they built in the codes. The paper ralph pointed to claims that it was a Richard Hammings.... perhaps this is something to be cleared up...

I seem to be sturuggling from Nash's problem of comming up with an original idea. I think it was Terry Pratchet in one of the Diskworld novels who asked the question "It takes a genius to invent something new, but how much of a genius does it take to invent something that allready exists?"

Anyways, perhaps Ill find a gap in the knowledge of humanity in these documents. And if I do, then perhaps .. just perhaps I can fill it a bit more.

1 Jul 2002 (updated 1 Jul 2002 at 21:34 UTC) »
ishamael: thats not quite how i meant "one off" i meant as like this

000 is one off from : 001, 010, 100
101 is one off from 001, 111, 100

So a single bit is has changed - or rather a single bit would not cause an error. If you where to have your data, then wrap that in a hamming (things like parity, checksum and so on,) these would be done on the data. THEN wrap all of that up in this system, should reduce the need for resending corrupted packets (assuming a reasonable amount of corruption).

I suppose my next task is to simulate this with random corruption to see the reliability. Of course the downside is you're sending three times as much data :(

sab39 as he correctly pointed out. Well I am working on it. Oewww..oh yes, i had thought of that, and I recognise that my solution would only work assuming that you didnt "loose" a bit, only that a bit could get manipulated. But i think that checksumming the data should be done inside this encoding, and that this solution would *NOT* replace checksumming, just try to reduce the need for the resubmission of packets (in reference, for application on the scale of something like the voyager probes, where for a request at retransmit would take a week or so to arrive there.... and then a week or so back (or a month ...) those satialites had somewhere in the region of 150 error correction bits in relation to one bit back.

ladypine, you have it more to what i was originally suggesting in the original problem. What if when setting up a network we could define how much distance between words do you want? perhaps on low error networks (physical connection) we can say 0 differnece or 1 difference (for speed) but perhaps on high distortian wireless, or long distance radio connection (i admit here i have an ameature radio link in my past) then you can say a distance of 3, 4 or 5... but i dont believe that I have sufficently solved my original problem generically enough to say this yet.

Interesting Problem Number one
I've been asked just why I am doing this, and indeed posting this in Advogato. Perhaps I didnt explain my reasons too well. You *could* take this and implement it as a protocol. - why?

Well, assuming that you wanted to send a sequence of "a"'s and "b"'s... if you say that an "a" is eqivilant to 000 and a "b" is is equivilant to 101 (this is using the three bit solution described on the 28th June).

The sequence that I wish to send is this:

(a = 1, b = 0, this is "hi" in ansi)

baababbb baababba

This would translate into the sequence

101000000101000101101101 101000000101000101101000

hokay, so perhaps I've just made the string three times longer then it needs to be... BUT.. if i change one bit.. lets say bit number five...

101001000101000101101101 101000000101000101101000
This would cause the computer some disconcern if using a standard transmission of a=1, b=0 BUT, in this case, it dosnt matter, because 001 is also linked to 000 (its one bit difference. The idea being, in each set of three, you can have a bit wrong, and there would be no corruption in the message. Of course, if you had two bits wrong in the bit, then you have corruption.

great huh?

I've also realised that it dosnt matter what base you are working in.

Work
Hrm So my webserver is down again. What do I pay these people for? Somebody please remind me why Square Internet exist?

28 Jun 2002 (updated 28 Jun 2002 at 18:49 UTC) »
mattr: I must admit that I rarley use the advogato thing to find out what strangers are doing, but being bored, I was following links. And I behold on a notice that you had certified me. Well thanks. But thanks more to the fact that you had said why. I think it was fxn (i think thats the guy) who recently said he didnt understand why people certified other people, or how they where to rank them. Well, perhaps this should start to be the start of a fad. Why do you give people the recomendations that they deserve?

Personally, I see myself as an apprentice. Of course this is by the standards of the people who use this. Compared to the average person on the street, they are unable to compare the difference between some of us. What I am interested in is bytesplit certified me as a Journeyer. Not that Im complaining to have lost my green colours or anything, but still...

However, I must object strongly to the accusation of doing computer science. I am not, nor ever will be, a computer scientist. Why? Well look at the people who are doing the computer science course? Half of them dont have a clue. Certainly, they have no ... feel for the computers. They dont seem to care about them. And because of that, the general "ComputerScience" course is becomming slowly usless. No, I am an Engineer, and even though we share certain modules with the computer science students, I feel that this course is a much better use of university time. Im sorry, but I really dont think that anyone with a Computer Science degree has actually done anything worthwhile. Now perhaps Im wrong about that, but after just finishing my first year at UEA, and not having covered anything that I hadnt covered at Alevel and in my own interests, I feel that I have aceademically wasted a year of my life.

However non-accademically... waaaahoooooo

Well Ive been having a little bit of problems with my machine in the last week, with hardware problems, taking hours to get past the POST stage.

Interesting problem number one
Actually, Ive noticed something, it dosnt matter what base you work in (assuming that the differnece change in bit is that the bit can change to any other bit not just to short range of bits either side of it... hrm didnt explain that well).

Its reasonable to say that on a base two system:

A b bit number will have b neighbours.

A b bit system of the nth base will have n^b combinations

So the maximum you could have if you ignore the blatently obvious fact that perhaps it just wont fit... is

max = (n^b) / (b + 1)

SOLUTIONS FOR THE FIRST FIVE BITS:

One bit

0 1

-O--+-

Two bits

0 0 1 1 0 1 0 1

-O--+--+--+-

Three Bits

0 0 1 1 0 1 0 1

| | | | 0 -O--+--+--+- | | | | 1 -+--+--O--+- | | | |

Four Bits

0 0 1 1 0 1 0 1

| | | | 00 -O--+--+--+- | | | | 01 -+--+--+--+- | | | | 10 -+--O--+--+- | | | | 11 -+--+--+--+- | | | |

Five Bits

(note that this is a 3d solution - the two graphs represent layers)

0 0 1 1 0 0 1 1 0 0 1 0 1 1 0 1 0 1

| | | | | | | | 00 -O--+--+--+- 00 -+--+--+--+- | | | | | | | | 01 -+--+--+--+- 01 -+--+--+--O- | | | | | | | | 10 -+--O--+--+- 10 -+--+--+--+- | | | | | | | | 11 -+--+--+--+- 11 -+--+--O--+- | | | | | | | |

Table of results:

+------+------+------------------+----------------+ | Base | Bits | Formula's answer | Found solution | +------+------+------------------+----------------+ | 2 | 1 | 1.00 | 1 | | 2 | 2 | 1.33 | 1 | | 2 | 3 | 2.00 | 2 | | 2 | 4 | 3.20 | 2 | | 2 | 5 | 5.33 | 4 | | | | | | +------+------+------------------+----------------+

Ok, so I had my first meeting with Peter Francon today. Suprisingly or not I've met him before - gliding pilot, hmm that takes me back a few years, actually to the year that I was in 14!!! wow.. five years have gone by since then.. sheesh I must have been an annoying goit.
No, but we've made some progress today. Hokay so im hacking PHP, but its paying the petrol. ahhh petrol...... ahem, back to supporting the greenpeace front i think.

22 Jun 2002 (updated 22 Jun 2002 at 15:27 UTC) »
Jobhunt
Status: Successfully gained freelance project work for a couple of people. Possibility of training.

Interesting problem number one:
The basis of this problem descends from an idea in Jack Cohen and Ian Stewarts book "the collapse of Chaos". The idea is this:

DNA contains some (if what limited) redundancy. This redudancy is in the form of the dna sequence CAT where to be mutated into CAC that they both specify the same amino acid.

Could this prinicple be applied to computer communications? To see, I proposed the following question:

Inital setup:

Using the nth base, and having a number which has a length of b bits, there are of course, b ^ n possible different permutations of the number (Assuming that you can not have any blank spaces).

These numbers can then all be placed as nodes on a grid, with lines connecting each point of the grid elsewhere, but they are placed in such a way that if you move from one node to the next node, only one bit will change. I.e. using a base ten system, you could make a valid move from 15 to any in the following set:

{ 05, 10, 11, 12, 13, 14, 16, 17, 18, 19, 25, 35, 45, 55, 65, 75, 85, 95 }

For a base 2 system, this is commonly known as a Kernaugh map.

ie 0 0 1 1 0 1 1 0 Where the number for each node can be read by | | | | reading the top and side values.. for the point 00 -+-+-+-+- marked with a O, it represents the value 1101. | | | | 01 -+-+-O-+- | | | | 11 -+-+-+-+- | | | | 10 -+-+-+-+- | | | |

Question:

How many points can you get onto the grid, where each point satisfies the following rule:

a point nor its direct neighbours (ie all the nodes which have a single bit's difference between them and the point) can not intersect with another point OR another point's direct neigbours.

Is there a rule to say how you should arrange them? What is the relationship between the number of points and the number of moves between them (assuming you can ignore the above rule).

I think that I have solved it for a base two system, but am working on a proof for it. hahaha

Erratum

A b bit system of the nth base will have n^b combinations
UUNN/Orgy
ok, now doing some sick idea of tying the two projects together (!) and also am doing some more UUNN programing in ahem perl... ok dont tell anyone im doing that :)

Life
Hmm things going interesting. My attempts at doing freelance work for some webdesigners are going down the drain fast as they all are out of work themselves. hehehe. Also I have found that the job center got shut down about five years ago!.. and Im SURE it was there last week.

14 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!