Interview with Kozmik/Nextempire

An Interviewing Prophet of Hugi


Introduce yourself to our readers - who are you, where do you live, what do you do?

I am KOZMIK. A 28-year-old half-French, half-Russian male. I lead the NEXTEMPIRE demogroup, where I also code. I started coding in 1981 on a Texas Instrument TI-99 and haven't stopped since. In 1983 I was sysop of the sKool BBS in Ottawa Canada (running off a floppy diskette drive), which became the perfect place for people from "pirate ship" of Hackerlite to meet, and I was recruited then. I guess that's when I started to be in contact also with various so-called "sceners" like DefJam and Quartex through Unknown and others. The movie WARGAMES hadn't even come out yet and you could still do everything you're doing today with the net. I graduated in Electrical and Computer Engineering specialized in Artificial Intelligence in 1996 in France. Mostly since out of school I basically have coded solely security related software as some people "do not believe in AI". Recently I have been recruited by French government still in the security field and have succeeded in reporting first day of work after Gathering'00 so I get a chance to finish an intro I ought to present there.


We're already looking forward to the intro. Can you tell us some details - what topic the intro will be about, for example?

I'd rather keep it a a secret and surprise the Gathering crowd. The truth now is that I am late, and I have no clue on what it will be about to be honest. At most I am still exploring 3 possible threads parrallely. I have some effects and I just finished "deciphering" the main effect of a very well known intro after watching it 18th times. The track is heavily based on vinyl "scratching" though and you can expect the subject and motion to be somewhat related to the track. My lack of time comes from the fact I had to pass several contests over a non-stop six months period to obtain the job mentioned here above, and it seemed of higher priority than anything else since it is garanteed for life and would in turn garantee for life the demogroup's financial existence and stability in the future, once and for all.


Artificial Intelligence is a very interesting topic. I remember a wise man saying: "If we knew how human beings thought, we could also say if it is possible to create thinking machines." Why do you believe in AI, and how do you think the AI research will evolve?

I believe in AI first as a field of studies, whether or not some of its sub-domains have plausible goals in the 21st century. When you study AI you learn that everybody has a different definition for it. It ranges from "making computers think" to anything not possible through the "traditionnal methods" and sometimes you just end up with Non Polynomial or Strongly Non Linear problems, where you mostly can only search for exhaustive states for now. Scanners "imitate" the human eye and automatic character recognition has proven to be effective. There is one debate to avoid though, which is : if something mimics to perfection some other thing, can the two be confused or is one just "a copy", knowing the fact that "mimicing" constitues the basis of human learning? The main problem to make computers "think" even years after Eliza, boils down to exactly the number of different mathematical combinations that can be expressed at one time on one or a set of computers. You have probably heard that if we "stretched" one's neuron tree, we could go to the moon and back, and magnetically storage-wise we still are far from that. Then there is the problem of overlooked stimuli "senses" like smell and touch. At last the maybe unsolvable problem in "making computer think" when you look at the BIG picture through neurophysiology is the electrical remanence wonder, which is briefly resumed as all your life memory and actions are pre-determinated by a single sigma electrical noise from previous refreshes (yes one knows nothing and life is some form of loop), a little bit like when you pull off the vacuum cleaner electrical socket, and the latter still runs for a split second. The problem is that at one point in the remanent loop before the dynamic mapping at t+1 of one's somatropic map (how one applies individual weights to neuron motors linking to brain cells to filter stimuli in order not to jeopardize one's knowledge at time t), collective consciousness comes into play physically, so it's a dead end except for philosophers or people able to predict the "Butterfly Effect". Anyway life is not hard to understand discreetly, but understanding it unvails void, and the simpler it gets, the more awed one is at it's wonder and that's really an understatement. Someone made life optimal.


Your group, Nextempire, is pretty new, but it already seems to be well known. Could you tell us a bit (or a lot ;)) about it, like what's the style and profile of the group, the group's history and projects?

Ok. First of all NEXTEMPIRE is about doing non-serious things seriously. It's not serious as is, and nobody thinks they are doing serious things, but doing them seriously. The demogroup was created a year ago, and was statistically the most productive new demogroup last year with around 20 demoparty/demomags releases. We still have yet a lot to learn and certainly a lot to improve including the quality of our productions. But can you picture a pitbull holding on to your arm? That's how strong our determination to evolve is. To do things still seriously, NEXTEMPIRE is a registred company and ought actually to be called NEXTEMPIRE INC. NEXTEMPIRE is also a registred trademark. This was not meant to head towards commercialism, but rather "lock" the demogroup's destiny.

Members come from all around the globe so there is no real "country" or "scene" domination, and that's one of the reasons we often neglect/refuse to state our nationalities in registrations, as it goes against this group's spirit. In short we are demomakers before being from this or that country. If it was the other way around, it would not make any sense to NE. Also some members have nothing to do with the "scene", but are talented in computing, and some came as "artists".

We mainly interact in English through emails and ICQ messages, and we have a mandatory once a week IRC meeting where we lay everything flat on the table and talk about future orientations. Sometimes newsletters are also required to cover short and confusing periods of time where a great amount of "little" things need to be done by many. Centralization is therefore of some importance regarding the above. It has been said by some there is pressure in the group to DO things (as in acts and not words); too much for some. Since we give a chance to anybody honest and polite enough, a lot of people come in but most of them leave because of pressure a month after.

The group has an infamous handle format for all its members, which is 6 letters in length and starting with the letter 'K'. Because of this we were fast to be coined as relating to KKK, which couldn't be further from the truth when you look at the various origins of the members. The main reason for this is intrinsic and tacite cohesion. A chain is as strong as it's weakest link. By changing your handle you are proving your commitment and trust. In that sense, we are very much like the first wave of demogroups, strict and no new-age "all opinions are equal" although we all feel this way, but from experience too much open-mindness is somewhat sometimes a brake to time-contrainted progress. Generally speaking, studies show one does 80% of the work in 20% of the time so we try to focus on the time/result ratio.

Our main day-to-day goal is to release at demoparties and eventually some diskmags. However general it may sound, it's actually very precise. At every demoparty we participate to we try to submit at least in intro/pixel and multi-channel entries. We also try to always have a party report and some photographs. Photographs are important not only for vanity ("All is just vanity and even if I write these words, it's still because of vanity" - Rousseau), but mostly because on one's deathbed, all one has is souvenirs.

We don't really have a style, (we'd like too maybe) but that has not come around yet. Our style is just to produce for now. It will come naturally with time. This is not to say we do not admire the early Amiga productions more than many of today's pc productions.


What are the current projects of Nextempire apart from the intro for TG00?

At TG00 we will compete as said above in pixel and multichannel categories also. Same for Mekka & Symposium 3 days later. Now in the future, as dull as it may sound we intend to compete at demoparties we think worthy of competing in (demoparties who demonstrate the minimum respect for contributors where the winning demo entry gets shown on the big screen as it has recently NOT been the case), and of course again progress in quality. Then we have mid-term as well as long-term projects. The mid-term projects include setting up Nextempire's Internet radio for our musicians, in order mainly to augment their chances of getting maybe a record contract if that is their wish. We will transfer our temporary website to our own DN1 which as already been reserved. We would like to augment the number of members. Long-term projects remain confidential to much of my dismay. Let me just state we have no plans of "englobing" anybody directly or indirectly, so there is no need to act upon such unjustified fears. Live and let live. Simply said, it is my opinion that there is no greater reward to have a demogroup outlive it's founder.


What should governments do in your opinion to prepare students for the modern technologies?

It is difficult for me to have any form of political opinion (political in this case means anything related to governments) considering my current status. You probably are alluding to education systems in general. There is little mystery about what needs to be done. There is a need for computing to stay as a specializing branch but also be integrated as a basic course, just as mathematics, physics/chemistry, litterature (national language), history/geography and economics. If the course was to last an hour (it has been proven one's attention span drastically decreases after that period of time), the first part of the course could focus on general computer related material (how to use MSWord, how to "surf" on the Internet safely etc...), and the latter basic programming teaching in BASIC language for example. I do not believe there is a need to "educate" people regarding the respect of individual privacies as it's tacite and therefore unfounded. People violating the law, whether through some modern medium or not, will always do it knowingly and ought to still be considered as "simple" criminals. On the other hand most of the education system is based upon the past in an authoritive fashion, whilst it should be turned towards the futur. More freedom and encouragement towards creativity will always have greater response from students. Furthermore problems encountered in the accomplishment of a personal project will undoubtfully bring most to the past's knowledge in a more involved and beneficial way (is there a faster way to know if this number is a prime? - Euclide).


Which advice would you give to people new to the scene who want to start coding demos themselves?

I think some people in the past have done more than just typing a few words to answer that question, so my advice to someone in that position would be to respectively:

1) Avoid IRC and newsgroups at first.

2) Read Trixter's definition of demos.

3) Go buy a beginner's book on graphics computing at local computer library and let it be your bedbook.

4) FOCUS on outputting a pixel at coordinate 160,100 in lowres mode, and do not let go until it's done.

5) Download all of Hornet's archive. Look at how other people thought of things differently. Do not be discouraged by things you do not grasp, it will come by itself, always focus on coding the "next" thing that is just a little above what you know how to do for sure.

6) Think of which language and platform you are aiming for (remember only 5% of the world population has access to Internet and 3d cards still in 2000). Do you aim for wider public or how someone said recently the "big bang" ?

7) Start going to demoparties to submit your demo/intro. Don't have the "blank page" or "perfection" complex. As most things in life, if you wait until you are ready, you will wait your whole life away. If your demo/intro does not make the cut at first (aka the jumbo screen) don't be disappointed. It gets easier as you go, but there needs to be deadlines, without them, you will not progress nearly as much. So accept them now. Do not confuse pressure (obligations) and passion (ie: you wanted to). Be objective and true to your apirations. Be strict to yourself for yourself.

8) Remember that mathematically SUCCESS=WORK*MOTIVATION.

9) Look for a group, one that suits your "state of mind" and pace. If somehow you make it in a well known group at start, there is little chance it will be an enjoyable experience.

10) Have a special directory on your HDD containing that one demo/intro that made you switch from coder to democoder. That's your "battery recharge" lifevest. As an example mine was "Dark Room" by Weasel and Strobo on Amiga, and is now "Stash" on PC in directory "inspire". Find your own.

11) Avoid cocaine and lsd at all costs if you plan to last. Cocaine is supposed to increase the frequency of neuromediators in synaptic gaps, but it only lasts a few minutes, costs very very much, and is very dangerous to your heart. Lsd is the only drug known to man for which psychically one is not garanteed to obtain a return ticket for.


What types of articles should a diskmag publish most in your opinion?

General Info is a must for me. It's part of the scene "history" (i.e: who did what where, who left which group for which one etc...), with emphasis and priority given to demoparty organizers (last minute changes of dates/rules /location). But then to me a scene diskmag ought to me revolve essentially around reviews of intros/demos and most certainly demoparty reports, which ought to represent in my opinion the core of any diskmag. Interviews are important to get to know the point of view of someone you only know by name or that you might only have caught a glimpse at at a demoparty. Come to think of it, pretty much everything you and other major diskmags do seems to be right on the mark. So all I can do is list in respective order what I read first:

1) Editorial.

2) Demoparty Reports.

3) Interviews

4) Scene Articles.

5) Intros/Demos Reviews.

6) Music/GFX Reviews (if presented at a party)

I usually avoid:

- Tutorials (all of them).

- Poetry corner (although some might enjoy it - can understand that).

- Coding corner (Wilby is an exception).

- Hardware reviews.


Do you want to tell our readers something particular?

Be as positive as you can be every day. Stay away from anybody that wants to put all the competition categories in "one" big bag, like accelerated/non accelerated, or intros and demos, or soon demoparties will become a "free" publicity stunt for commercial game companies. If you have not gone yet to a major demoparty, you have missed a lot IMHO, so stop hesitating, grab a few friends and go there! Remember also it is very difficult to organize such parties, so try dismissing "minor" problems there and focus on "the happening". Respect your group despite the fact you will always encounter people for whom nothing is sacred. Always trust your "gut" feeling when in doubt.


Kozmik & Adok - 23 Mar 2000