Theo Verelst Diary Page

Latest: 21 december 2000

I've decided after good example to write some diary pages with toughts and events.

Oh, in case anybody fails to understand, I'd like to remind them that these pages are copyrighted, and that everything found here may not be redistributed in any other way then over this direct link without my prior consent. That includes familiy, christianity, and othercheats. The simple reason is that it may well by that some people have been ill informed because they've spread illegal 'copies' of mymaterials even with modifications. Apart from my moral judgement, that is illegal, and will be treated as such by me. Make as many references to these pages as you like, make hardcopies, but only of the whole page, including the html-references, and without changing a iota or tittel...

And if not? I won't hesitate to use legal means to correct wrong that may be done otherwise. And I am serious. I usually am. I'm not sure I could get 'attempt to grave emotional assault' out of it, but infrigement on copyright rules is serious enough. And Jesus called upon us to respect the authorities of state, so christians would of course never do such a thing. Lying, imagine that.

Previous Diary Entries

Dec 21, 2000

It's early morning, time to keep up with the latest diary update rate, as long as materials make it worth while.


I must a missing/wrong link in my diary chain, if I have the chance I'll throw in a list of all of em with a short index. My therapy? No, not realy, it was after an example that I like to read that I started writing, and it is sortof to make clear to who wants to know what some days of thinking and doing have had to offer, and occasionally to be reminded of things I took the effort of stating decent enough myself.

Considering a lot that's going on for me behind the screens, I won't comment on all subjects, but I'm active still in the job acquistion process, hopefully with sucess.

I'm meeting someone with possible job offers (in software) today, while I'm waiting for some other things in the same area. I've been offered (over repeated email) to do jobs in where I've been active in the past: courses. I prefer a good job in a programming type of function for the moment, that seems more to the point, and courses are often irregular and require different attention.
It's kind of akward: with my experience I've spent 2 months or so already wanting to have a well enough paid, start straight away software job for limited time span. One general idea seems to be that not too long projects are not on offering much, many simply are asking for permanent position to be filled. Donno why exactly, some freelance type of arrangements per company wouldn't do bad attracting some varied knowledge and skills. Not that I would want that in general or at the moment when I have an opportunity that might seriously grow on me, than I prefer a long(er) term possibility and outlook, but about other possibilities I'm just talking about using my knowledge / skills to generate some funds on do some job and get paid basis.

Programming again

There's lots of Java demand, I've even read up on IDL, and some other things (I have Java programming experience, and lots of OO exp with Objective C), being reminded of my work on distributed computing interface definitions, where I made proceses on different types of connected machines communicate and transfer (scientific) numerical data and data structures. Interface Design Langauge aims at simular things, calling it distributed objects (an object 'officially' being nothing else but a collection of program and data information in one unit or structure), and IDL is a JAVA implemented stub (communication interface layer) for CORBA defined interactions. Interesting to see that simply writing down the interprocess behaviour (message-wise), and having some method to convert daya between machine types, in fact suffices to sum the whole thing up, communication-technical wise.

Just like I thought about 5 years back or so the next (NeXT?) thing is to somehow program the virtual, distribued machine in a general enough way and language, preferably with automation all over. At the time on workstations thw former job would be done for instance by using standard packages such as RPC (remote procedure calls), just like remote object invokations under Java without using threads not usefull as distributed computing paradigm because when the other program is running, the caller is stopped (than again who cares much about efficiencu in Java, generally..), but it does function parameter packing over a socket stream, and it uses stubs between machine types to convert between variable types for different machines (the number of bytes mantissa and exponent, and their format and order for floating point numbers, for instance), so one can program without knowing how the code is going to get linked together in the end, that is which procedure in the end runs on which machine in the network. Disadvantages are mainly that the program may become distributed (running on more than one machine), but not parallel (they perform computations taking turns, not in parallel, at least not without using extra self-programmed processes and extra communication), and using the stubs around functions requires quite some extra compilation steps, though they could be automated.

Another way to arrive at programs that are generally distruable AND parallel was (and is) to use some package that basically does the jobs of data conversion, packing data into a stream, and interfacing communication routines in a programmer friendly way. At the time I tested PVM (parallel virtual machine), Isis from Cornell Un., with the heavy and interesting group concept, and some lower level desirablilities such as utilizing (ethernet) broadcast facilities for efficiency), and I wrote my own software, and even had someone else put it to use for the graphics project I was in, which offered the same functionality as I mentioned but on names-link basis, with efficient parallel behaviour based on a set of routines to make multiple connections set up over my own connection server, and their handling based on stream handling and select-call encapsulation library.

A main theme being flow control, which is not easily formally defineable, though it is possible, that's I've looked into CCS / CSP / SDL communicating process theory quite a bit, because that's where most bugs, deadlocks, and gross inefficiences find their source. Hardly any OS / env specifies all that is needed to know here is my experience, though tracking the Linux kernel may be an option to find out whats there.

Cisco c.s.

As many know (I didn't until some years ago) about 80 % of the networking machines in the world is filled with Cisco machines, to do hubbing (connection local area networks together as if one), switching (connecting networks up with the possibility to decide where datapackages go when they reach the boundaries of a network section), and routing (do the main interconnection routing thing, make trafic go from one destination, over in betweens to its final one).

Connect to the internet, surf to some not-local destination, and you have a fat chance that some Cisco 19 inchers or whatever they look like at some point takes your bits and puts them on the next set of wires connecting you.

I've been reading on the fundamentals of the current technology in all this recently, interesting enough to know, and definately professionally important. Pity that gaining experience with let say backbone related equipment is hard to get, but then again, the theory together with my programming and networking knowledge (physical and at application layers) can lead to pretty good knowledge.

The main lets say general thought about the whole subject in sort of look-at-the-future way is to think about the idea of broadcast subnetworks, a concept that can be used to spread information efficiently to a lot of users. Lets say that I want to listen to internet radio, as I did some time aga, then the main advantage is: I can take any station I like! For instance netradio has nice 70', blues, jazz channels, waiting to be listened to over acceptable realplayer quality, sort of boosted AM through doably FM radio reception equivalent, depending on available channel bandwidth. Quite niceto work by, and the songlist is instantly available even with one click away album/artist/buy the record possibilities. What more could one ask for.

Now the point is that when I connect to that station, data has to be fed half way around the globe to reach me, under the ocean (or over a sattelite delay?), over my countries backbone, to my provider, and finally over my modem. The same as surfing, except its a continuous data flow. Now suppose my neigbour want to listen to the same channel (way it goes sometimes), then just as with surfing the web, normally he would get his own connection to the same musci data server, get his data transfered over the same or different internet route, and gets his audio (or video) connection.

Now suppose Madonna gives a concert (not that thats imo a particularly desirable occasion, but still, completely legal), and all over the world people want to 'tune in'. Massive data transfer would occur from the server with the concert feeds. When 100.000 people want to connect up, a hundred thousand links would have to be made, which even for heavy server workstation with Gigabyts of memory, hundreds of fast pentiums equivalent horsepower and lets say actually utilisable full ATM connections would be not doable.

There is a provision in netland which makes all this possible, I'm not sure how much it used, and what the implementation details are, but it is possible to define sub-nets. That's sort of a name for a collection of connections, which are of a broadcast, that is one transmitter, many receiver, kind. Suppose the concert server connects up with all subnets that contain listeners, and makes sure these subnets contain amoung others data units that contain the concerts audio/video material. Then what is needed is to make sure all those subnets get their feed, and that somehow I as 'member' can pick my data from it. That means that my neighbour and I probably draw fromthe same data stream, when we have the same provider. And otherwise, data probably has to pass the ocean only once, to be transfered to .nl and the providers.

I know some more about the possible techinicalities, also network-wise (I read on it some time ago), and its an interesting subject imo not at all implemented to any general and complete level.

Can I have this over a sattelite link in my car (if I have one again)?

The Art of Noise

Securing data is probably the most practical idea. I'm certain that an electronical aid may at some point be very valid in all this. That device would be a noise generator, because I'm certain that lets say zener diode (reverse base emitter transistor junction) or straight diode noise is random enough to realy be unpredictable. Add a decent enough analog to digital converter, and a little library, and one can generate random numbers by the million which are absolutely unpredictable. Quite completely even when needed. Good for generating keys and match data in encryption algoritms.

I've seen my countries main electronics magazine present USB single chip solutions, that may well be used to give all this a general application side. I shouldn't give my ideas away maybe, but then again, all this takes considerable knowledge to implement right.


Who is M. Zucker? Apart form the phonic associations, extra info is needed: it is a historian who in '45 published 'philosophical perspective of american history' (or something to that extend), two parts, one about field theory (as in physics) in historic sense (as opposed to gestalt theory?), and who after a number of pages reading captures my interest. I tried to look up who that person actually is, contrary to good habit there is nothing about the author in the two books, and both the library index and the library of congress index have nothing to say about the man, which is sort of insecurifying about the authority that can be ascribed to his writing, but it is interesting. Normally it would read 'mr so and so was a respected fellow os university so-and-so, has interests in this-and-that, publications can be found on so and so' and the preface would contain the obligatory but explanatory thank yous and descriptions that give an idea of a persons background and interests.

For instance I was reading about the time of crisis, and the fact that people were affraid in about 33 (or was it '29?) when the bank world was maybe going to be state business instead of private, and rooseveld I think it was was the main player to make things happen against for instance a congress that he (m that is) describes as 'without will', and such. In other words mass communcation hits the world (spiritual ?) system, money is at stake, germany is a factor to play devide and conquer politics in the US, and the whole thing blows up.

Some other analysises are equally interesting, I'll make some notes while reading, that should make these texts better.

NYRB has interests me, thats clear enough from my pages, another reason is the sort of university idea in it, where major universities in the US and brittain (rarely others) make their points about what they think is interesting reading nowadays by making adds with nice book pictures and titles. Various ones at least make a lot of sense to me, and indicate that my thinking may not necessarily be hip, but definately has correlation with theirs, which I consider positive. I'll make notes to clarify.

Revelation has another few verses explained in my mind quite a bit better since yesterday, the part where the various types of believer communities are portrayed (Rev 2), sais something about people being knowledgeable about the 'depts of satan', where of course there is no need or desireability in that direction whatsoever, where the word 'depth' means also width (of a coastline), height, profoundness, etc, and that those who do not have such a doctrine (the d word), get no other burden from Jesus than to hold onto what they have until He returns. Now that verse is definately not just interpretable that way, it could easily mean that He casts no furter weight on them, and exhorts them to strongly hold onto what some/they have, until and/or as far as it has become or come into existence.

Interestingly enough I was reminded about various other words trough one Aeschius quote, about man/husband, a woman houshold leader, and related concepts, it a.o. rang bells about a construction in James 1 (where I failed to funished a section on in the previous page, that was mainly intended to go into the receiving of those around the ones of the Lord, and the taking of faces as prinples, and the monopod and distinguishing based on riches), talking about a man that is like a wave of the sea.

Phrase I read the other day: graciously obnoxious. Or gracefully obnoxious? Not the same as hidden agenda planned irritable. Not at all. That would give Cleese type irritations.