WOC Classic Q

[rocky]

gross, you're right:

6. Jury
Björn Persson (SWE), (IOF Event Adviser, Chairman of Jury)
Vincent Frey (FRA)
Ian Bratt (RSA)
Marek Petrivalsky (CZE)
Unni Strand Karlsen (NOR)
David May/David Rosen (GBR)
David May will be on the jury until Wednesday. David Rosen will be on the jury for Friday and Saturday.

so david may was on the jury yesterday.

[Adventure Racer]

I had a mispunch at a recent event, where I'm 100% sure that I heard a beep when I punched (I actually went back and re-punched as I think I was too fast the first time, though somebody else punched in between my two attempts). When I protested I was told that what I was suggesting was impossible. It's nice to know that the weight of evidence suggests it is possible I'm not imagining things.

I can understand the idea behind not interrogating the boxes to resolve protests, as it potentially unduly delays issuing results of competitions. But that is based upon the idea that the system works perfectly, when the weight of evidence seems to suggest that isn't necessarily the case.

Oh, and my incident was with a V5 card - have now upgraded to a V6 (though actually had the order in before this event due to a mispunch that was my fault for punching too fast earlier in the year), though not actually used it in anger yet. Presumably Oli has a V6 though, so it seems that may not solve the problem.

[FatBoy]

However, has anyone actually ever tried to punch without getting a beep and a flash with a new card?
You have to have reflexes that would be better used as an international sprinter, not an international orienteer!

Yes I have managed to do this in testing - to deliberately try and write crap to the card/unit - it is difficult though you are correct. I've never seen a time written to the card but no control code though. You see it plenty where a card number but no time is written to the unit, and a bit less when the unit is written but not the card. I've told many a competitor at an event (not necessarily orienteering) that if it did beep then it would be on the card, however I would trust Heather and Oli more than your average punter - they've probably had 1000's of punches more for practise! It sounds like this is maybe a bug in the firmware somewhere - it's beeped but failed to write the card number alongside the time. However some sympathy for SI because unless it is happening very regularly you can see why they would say it can't have beeped. Fixing an unreproducable bug is very difficult.

I think if the time is in the unit it should be allowed to stand at this level. Orienteering should be about navigating point to point in the shortest time not worrying about the complexities of registering with hardware. I'm sure SI would argue that their backup system has worked correctly in this case...

Gross and other diehard Emit fans - I'm afraid I know of at least one person who's been DQd because of (hire) card failure. The backup pin punches - great you can tell you've been to them all but what time did you take? For the record I'm not knocking Emit - just saying there is occasional failures here too.

[guessed]

The fundamental flaw here is that SI decided to record what has happened in the control unit itself. Doubtless they saw such a mechanism as being needed to match Emit's backup paper slip.

Either the organisers should be forced to fetch in the control unit and query the record, or SI should produce control units that don't record !

There was a lot of "mis-punching" in the early days of SI, and organisers simply did not want to be in a cold field for hours waiting for all the controls to be collected to prove to the competitor that nothing had registered. As I recall, that was the prime motivation for the rule.

I simply do not understand the need to get equipment with faster and faster punching times. Much better a delay of 1 second before the flash, so the hardware can add another read card cycle to verify punch is on the card. It is the same for everyone, whether it is 0.2 sec or 1.0 sec !

[Ed]

There's various reports of strange happenings with SI on a thread in the German O Discussion forum (see http://www.orientierungslauf.de/content.php?m=2&s=5&x=2&mid=12988). ). Particularly interesting is this post where Stephan Krämer (Mr Sport Ident) is quoted as knowing of problems in reading SI cards with the reader unit in "fast mode" (the results were showing that runners had visited non-existant controls two hours after they'd finished!). In most cases deleting the result and then re-reading the SI card corrected the problem!

To me it suggests that the data transfer in SI is not foolproof.

Given the use of video referees, third umpires, etc., etc. in major events for other sports then I'd say orienteering needs to revisit this rule. There is evidence gathered by the official time recording/punching system that a competitor has visited the control (and even the time is recorded so taking out of order etc. is not an issue - and if it comes from the official system its not relying on someones photo etc.) and the rules are to ignore that evidence !

OK, for minor events the cost/benefit of checking boxes may be too high, but for major events I wouldn't have thought so. Not every village cricket match needs to have a third umpire![/url]

[FatBoy]

Stephan Krämer (Mr Sport Ident)

Just for the record Stephan is Sport Software (OE200x etc), not SPORTident. I presume "fast mode" he is talking about the new commands for hardware versions >= 7, and I know he is not a big fan of those. Personally I find the new commands far more reliable than the old ones.

I would read the thread but my German really isn't that good!

[Ed]

ok, Mr Sport Software. Just having scanned through all comments on the thread again there is also a case described where a card was sporadically registering punches in the unit without error codes but nothing on the chip itself (in one competition five "no punches", two of which the organisers out of interest checked and then decided to reinstate the competitor as they decided it appeared to be a technical malfunction.)..

Krämer had said he didn't believe it was a faulty card. The card got sent back to SportIdent and a while later there's another post saying he now does believe in faulty cards, but that an explanation from the manufacturer wouldn't be forthcoming.

It would be interesting to know if anyone has ever read out the contents of some of these other boxes to see whether there's error codes recorded with the punches: and whether anyone would blow the whistle if the punches were recorded in the boxes cleanly... which could be v. embarrassing for SportIdent.

[FatBoy]

It is fairly normal to read a station and get punches registered with either no time (which shows up I think as the two main time bytes being EE or FF - can't remember which but results in a time > 12 hours). If you read these in SiManager or SiConfig they show as "err" for the time.

It is also normal to find punches where there wasn't a time written to the card - you wouldn't see these without looking at the card though obviously so not so easy to pick up.

However in both these situations you wouldn't have expected for the unit to have beep/flashed as it hasn't finished writing. Mostly if not always where you see a punch which either doesn't have a time, or one that doesn't correspond to the downloaded cards, you will then see a subsequent punch where the competitor has put the card back in. All perfectly logical.

The situation with Oli's card where the time but not the control code has been written to the card is different though. My knowledge of the punching process is by experience rather than knowing what order things are written in, but I've never seen this in my experience - leading me to believe it's a bug in the firmware or faulty card or something else.

[EriOL]

Whichever moderator removed the words
You Prick
from my last post can put them back

If Eriol Is going to brand my team mates and friends as liars then I am going to call him or her a Prick

thanks very much

either that or you can remove Eriols post which I find wholly offensive

I'm not calling anyone a liar. It's just a matter of very simple psychology. When you have punched thousands of SI-controls and the beep and flash has always had a delay of about 0.2 seconds you get so used to that scenario that you simply wont notice if the beep and flash doesn't go off. If I had to choose I would always trust an SI-unit over a human mind, even if it was my own mind.

[Adventure Racer]

If I had to choose I would always trust an SI-unit over a human mind, even if it was my own mind.

There speaks a man who doesn't work with computers

[little slave]

but you know that normally every ship is registered in the si-boxes...so normally they are able to control it like this....but now its too late

[Toni]

I'm absolutely amazed at the decision if there is a time registered on his SI card - I can understand the decision if (and I wish someone would clarify this) it was a radio control, Oli got a time but it didn't register on his card. Then they can explain it away through his punching - I'm not saying it is fair but I can understand how the jury justified their decision as taking the radio result would be the same as interrogating the unit. I have never heard of it on someone's splits but without a code and if this is a new thing then surely the jury need to be seriously called to rethink this.

It makes a joke of the sport - particularly where they have all this media attention and debates surrounding technology mishaps do the sport no good. In WOC with so few athletes and so few controls surely it can't be that big an issue to interrogate the units and this must be the only way forward - perhaps nopesport should start a campaign prior to the next WOC! Collect signatures, get someone to write an argument and submit it to IOF etc....

Also I think it is an issue if radio controls are slower as I am sure that subconsciously we all get into a rhythm where we know how long punching takes and then throwing in radio controls that take longer and throw the rhythm out seems completely unfair.

Quote Kitch

because these people are not 100% committed, 100% focussed, 100% on the ball - they just bumble round the course in a day-dream, it is only the World Championships after all.

Nice idea Kitch but (just as I think EirOL should not call Oli's explanation into question as she wasn't there) it is probably unfair to imply that the numerous athlete's DQ'ed at WOC were bumbling around the course in a day-dream. But Kitch - I can understand the patriotism and your reaction - I would be exactly the same - it's really quite nice to see and I'm sure it's nice for the GB team to see everyone behind them!

[DesignatedDriver]

i am afraid i don't know much about the writing process for the SI units - i guess:

Write Data
Verify
Repeat if necessary
Return OK

presumably if there was a fault in the verification algorithm e.g. at 13:27:45.89 the verification returns OK even if it isn't. Even the best testers in the world aren't going to be able to check this. but think about the number of controls at each event, punched hundreds (or even thousands) of times. There is going to be a point when someone punches at exactly the right (or wrong) time to cause this fault.

testing in a backroom, by the guy that designed it is one thing, but its not truely tested until it has run in the real world without problems. Can SI really be SO confident in their abilities to be able to say there is no problem? Sounds a little arrogant to me.

[Adventure Racer]

Do radio controls take longer because they are based on older versions of the control box (which the ones I've seen have been) or is there some other reason?

[Kitch]

Toni
we are missing a 'heavy sarcasm' emoticon.