[] [PvdH] [Home] [Newsletters] [Top]

CSS features best viewed with Microsoft's IE 3.02 or 4.01 or Netscape Communicator 4.05

Newsletter 12

Last modified: 5-Jul-99

KiteFlight & AJSKA/AKA/STACK

The topic is a PC program which I developed in 1993-1996 called KiteFlight. In short, my friend Daniel Rohner and I are basically volunteering to

• modernize KiteFlight's user interface (create a Windows 95/98/NT version),
• make essential enhancements and changes (but we are VERY busy people, believe it or not),
• eventually make the source code publicly available (unrestricted use?),
• try to make the software at least as reliable as commercial software (I believe we managed to achieve that in previous releases; it might be more difficult this time for technical reasons), and
• do all this work essentially for free.

• we are not expected to grant everyone every imaginable wish (we are relatively sane and know that we cannot make everybody perfectly happy; we don't believe in designing software by committee - we decide for ourselves what we think is the best approach),
• people understand that this make take quite some time to achieve, and
• the software will be put to really good use.

What's KiteFlight?

I developed KiteFlight in 1993-1996. It is basically an MS-DOS program which also runs well under Windows 95/98. KiteFlight and its upgrades (available via the Internet) included a relatively complete collection of animated STACK compulsories as well as various routines submitted by users. See the KiteFlight home page for more info and pictures which show what the program looks like.

The business side of KiteFlight

The price of the Kite 2.0 software package should be around $15=13 € (your luck may vary). The shops, distributors and publisher made (some) money. The software developers, Peter Ruinard and I, received a few percent of the sales price each - meaning that my KiteFlight income was comparable to my expenses (software development tools). This wasn't a real problem for me: I didn't "quit my day job", and never expected to earn any real money this way. The total number of copies sold (about 2000) was pretty impressive. So, in short, KiteFlight never really was a commercial product from my perspective: I was satisfied as long as it had enough content and enthusiastic users.

Developed by devoted amateurs

What is KiteFlight used for?

All-in-all the users can probably be classified as:

• persons who used the diskette for one day only ("looks nice" - probably the largest group)
• persons who bought the diskette in order to use Peter Ruinard's template printing program
• persons who used KiteFlight to view compulsory routines (e.g. T04.kf1)
• persons who studied the available full-length routines (e.g. Bluewing.kf1)
• persons who developed their own short routines (reasonably easy)
• persons who developed their own full-length routines (significant amount of work - probably the smallest group)

KiteFlight wish list

KiteFlight questionnaire results (Dec. '98)

Priority	Requested improvement	Score
#1	make a real Windows version of KiteFlight	46 points
#2	make it easier to create your own routines	42 points
#3	add ability to directly print diagrams	30 points
#4	add ability to calculate/display entangled kite lines	26 points
#5	add support for axels, etc.	25 points
#6	add 3-D graphics	16 points
#7	add ability to start animation at any segment	16 points
#8	add support for quad-line kites	13 points
#9	get STACK/AKA/etc support for KF1 standard	10 points

Note that the questionnaire was presumably mainly answered by relatively serious users.

KiteFlight for Windows?

The work is progressing slowly, but initial results are promising. Daniel's plan is to first convert the existing KiteFlight 2.0 and concentrate on getting a standard Windows user interface. And then to gradually add modifications and features. Note that Daniel and I are both very busy people - so the work is going slowly. The new version (code named KFW32 or KiteFlight 3.0) would solve at least questionnaire priorities #1, #3 and #7 and should provide a solid basis for solving some of the other requests in the future. We will present our ideas for solving request #2 (ease of programming) in the near future (it has high priority with us as well as the users).

Currently the KFW32 software can:

• load KF1 control files from hard disk (and save them again)
• allow the user to edit KF1 control files within the program
• display the text of KF1 control files (New: automatic text coloring! In the picture below, comments are in green, instructions are in blue, etc. Note that one instruction contains a typing error and is thus unrecognized: black instead of blue.)
  
  
  
  
• check whether a KF1 file is syntactically correct (see error message about line 10 in above screen shot)
• animate KF1 files (incomplete; see picture of animation screen below)
  
  
  
  
• allow the user to pause/continue an animation (new feature! see small Stop/Pause/Start control buttons in animation screen shot above)
• allow the user to adapt the color schemes of both the editor and the animation screens

We haven't worried at all yet about how we will distribute it (CD-ROM, Internet downloading, E-mail, etc). We also haven't worried about pricing yet: from our perspective, the software can be provided for free, but we may need to charge something depending on the way the software is distributed. KiteFlight 2.0 cost something because it was distributed through shops, and strangely enough publishers, distributors, shops and CD-ROM services do need to make money.

Note also that particularly the screen shot of the animation just reflects what it happens to look like at the moment (I may replace it once in a while).

STACK compulsory routines

Compulsory routines on the Internet

Resource	Description	Number of routines ( I+P+T+qI+qP+qT )	Main features
link	KiteFlight Compulsory library by Simo Salanne	47+12+31+0+0+0=90	Color animations. Downloadable. Require KiteFlight viewer. KF is also authoring tool.
link	Official STACK site with routines by Hans Jansen op de Haar	23+15+19+12+3+0=72	B&W GIF bitmaps. Viewable online. Require Internet access. View-only.
link	STACK Luxembourg by Gregor Westemeier and others	47+15+36+12+3+0=113	B&W GIF bitmaps. Various animated GIFs. Viewable online. Require Internet access. View-only.
link	Roy Reed's STACK animations	48+15+36+12+3+0=114	Color animations. Viewable online. Require Flash plug-in. Require Internet access. View-only.
link	Personal site of Hans Jansen op de Haar (ex-president of STACK)	28+11+22+12+10+5=78 "'Old' figures that are -almost- impossible to judge good or fairly, or give an unfair advantage to small or large teams, or to odd or even numbered teams are left out!"	B&W GIF bitmaps. Viewable online. Require Internet access. View-only.

The above table summarizes the libraries of compulsory routines available on the Internet. As the list is clearly biased towards STACK and Europe, it is safe to assume that there exist other sites which I haven't found yet.

KiteFlight is incidentally the first animation solution to become publicly available and the first to be distributed via the Internet. Some entries in the table provide (static) pictures only while others provide (dynamic) animations.

From a user's perspective, the view-only solutions are probably easiest to use. KiteFlight is also pretty easy to operate as a viewer, but has the most options (e.g. animation speed). KiteFlight can handle small (e.g. 5 second) animations, but also full routines (e.g. 5 minute animations) effectively.

KiteFlight is also different from the others mentioned above because it provides more than just a collection of "pre-cooked" pictures or animations: you can adapt animations and create new ones for personal use or for distribution. You can also analyze the motions by looking at the control information (e.g. see that a speed of flight increased by 20%). The other entries in the table (GIF bitmaps, Flash, animated GIFs) were produced either with professional design tools (e.g. AutoCAD or Macromedia's Flash authoring environment) and/or were created using relatively complex transformation tools/steps (e.g. awk scripts). In both cases, significant software experience is required to modify or make new diagrams or animations. Probably only Roy Reed's Flash-based solution is probably comparable to KiteFlight from an ease-of-development perspective (similar effort to learn the tool, similar effort to create a short animation).

KiteFlight <> AJSKA/AKA/STACK collaboration?

• complicated internal communication within- and between these organizations,
• the fact that that I only attended a few major kiting events (as a visitor),
• the fact that PCs and Internet were not as common a few years ago as they are now, and
• maybe even some technical limitations of KiteFlight itself.

As my last attempts to achieve collaboration date back a few years, and various things have changed since then, it seems an appropriate time to try again. One promising bit of news which has prompted this (shall we say) unbounded optimism is the fact that the Japanese, US and Europe/other officials seem to have started collaborating across their organizational boundaries. I don't know who to thank for that (I am an un-politician), but this might be the start of a brighter future for this sport in general.

Here are a few examples of the types of potential benefits of such a collaboration between AJSKA/AKA/STACK and KiteFlight plus the types of issues which could be tackled by combining our efforts:

• Unless there was a fresh edition of the rule book (with compulsories), it was not always clear to me who decides on which compulsories are official, what their description is and where the latest information was located.
• A compulsory routine encoded for KiteFlight is basically a supplement to a textual description. It is easier to understand - especially if 2 or more kites are involved - due to subtleties of timing. As a KiteFlight animation of a routine is much less ambiguous than static drawings (even with carefully worded descriptive text), it was not always certain if the KF1-encoded routines were "as intended". Note that there have been few (if any) requests to modify Simo's library of KF1 compulsory routines.
• Who to talk to? At one point Simo Salanne (for STACK) and Marty Sasaki (for AKA) were the coordinators regarding "Kites and computing" and we had lively discussions on what the software should look like and matters like commercial-versus-public, etc. The main problem was, however, that they were in practice helpful to suggest things, but unable to solve many of the AJSKA/AKA/STACK related problems. This is incidentally about 5 years ago now, so lets hope that organizational matters have improved significantly.
• The KF1-encoded routines have no official status: no STACK official double checked the resulting animations to determine whether the animations were as intended. Paper versions (diagram + text) were the official source. The KiteFlight versions were simply attempts to accurately capture what the we (Simo and I) understood from the official source. Matters were further complicated by important sounding matters like 'copyright' and control over the information. Not much attention was paid to setting up simple and effective procedures given the scarce resources (networks of willing, but busy volunteers like myself).
• Although KiteFlight adopted the existing AJSKA/AKA/STACK grid conventions (-100%..+100%, 0..100%), there was no formally accepted mapping of this 2-D notation to the 3-D motions of stunt kites. This problem closely resembles the problem which map makers have: the world is (shocking!) a sphere which must be projected onto 2-D sheets of paper. The 2-D representations work well for small movements along the surface of the sphere. Kites, however, fly large (e.g -50' .. + 50') distances on the sphere. Hans Jansen op de Haar has made a reasonable-sounding proposals to solve this ambiguity (e.g. what does "fly up" or "fly a square" exactly mean?). It basically comes down to interpreting the traditional X and Y grid coordinates as two angles in spherical coordinates (kites fly on surface of a sphere). I am interesting in (somebody) exploring this further: what if someone built an animator which really showed the 3-D motions of kites in this way. Would that be correct? Is the coordinate system internally consistent? Would that be useful, or is the real mapping between grid and real world more complicated than Hans' proposal? Can it be extended to display indoor routines (in which kite motions cover half a sphere rather than maybe 40% of half a sphere)?

Questions to the AJSKA/AKA/STACK community

• Q1: How much interest is there in a "new KiteFlight" (or something more or less like it)? It is a lot of work for Daniel (and to a lesser degree for myself) so I feel we "have a right to know".
• Q2: What can we (Daniel and I) do to increase the chances that the program will reach a sufficiently large and above all sufficiently active user community? As explained above, we don't need/want money for this, but some degree of recognition would be nice, would be appropriate and would be fair.
• Q3: Are there new/future developments within AJSKA/AKA/STACK on the grid and standards issues? Is the explanation of the rectangular grid at http://www.multinet.de/stack/compulsories/compgrid.htm official? Has anyone actually done something with it (e.g. for judging or practicing routines)? How far will that mapping from the "old" 2-D grid to 3-D reality take us? We can probably help with the technical sides (both of us are engineers), but there are also matters of requirements, policy, acceptance and vision involved. This point overlaps with...
• Q4: Are there things we can do to help AJSKA/AKA/STACK on the grid and standards issues? Examples: provide multiple-viewpoints of a routine, create a well-defined mathematical model of 2-D to 3-D mapping, provide a way of generating 2-D diagrams on paper from a KF1 description. If necessary with conversion between multiple input or output formats (computers are good at that)!
• Q5: Is AJSKA/AKA/STACK willing to support this activity in any way? Not financially (having this done commercially would require at least $100,000), but in terms of commitment, links to the rule book, advice, publicity, organization of the AJSKA/AKA/STACK side of things (e.g. compulsories), validation of encoded routines?

Direct any remarks or questions to Peter van den Hamer (vdhamer@msn.com).

PS. My thanks to the ten or so persons who have responded to my earlier text on the STACK-info mailing list. Many of your inputs are incorporated in this version.