Discussion: Opinions about e-punching in orienteering

Copy paste from a training log for a wider audience:

I'll try to write here some notes about e-punch in orienteering. Let's first try to figure out do we really have a problem or not. If we have, what are the problems.

Here in Finland we have used e-punch for about ten years now. You would think it's the only punch methods used, but still there is events with pin punch or without any punch. For example all do not know any permanent courses with e-punch, at least no any urban ones (there may be in one in Idre Fjäll). Also winter street O series here in Helsinki does not use any punch at all - there is controls and they take times and all runners do own Emit a brick, but still no e-punching is used. Why?

Answer is simple. Punch units are too expensive to be used as permanent controls or used in urban environment. Vandals break and steal them. If punch units would cost "nothing" or just a couple of cents/euros, it could be used for all races. For is it could be just a tiny stickers, you could put it on a control flag in this kind of small training events.

It takes a lot of work to program punch units and make sure . Those are now also quite big and heavy you can not set up training with 20 control just by putting e-punch units in your pocket and drop them to right places while running to course as training. We could have
units we don't need to program at all or worry about batteries. With small, thin and light units you could do it easily. Think about the amount of all the unnecessary volunteer work coaches need to do at the moment around the world, its huge.

This all does not mean we have a problem, but it do mean some things could be done in much better way.


(end of part one)


At the moment e-punch users seems to be relatively satisfied with the systems they are used to use. But when they have to use "the other" system, they hate it, because it's too complicated to use. SI users dislike punching with Emit, brick goes in only in one way and it's too painful and difficult to twist it in like that. If you are not used to it, it will
need al lot of concentration. No-none can be all the time 100% focused, and if you loose the focus while punching with Emit, you will not get the punch and you'll be disqualified. And it is the same thing the other way. Emit users are not used to wait the signal. Usually there is a huge noise at the last control you can not hear the beep or be sure is the beep coming from your unit. You have sweat in your eyes and sun shines brightly. If you are not 100% focused, you easily thing the beep from some other unit is from your unit an you go on. And get DQ.

This video tells a lot. Here is used SI, but the same can happen and has happened many times with Emit.


http://www.youtube.com/watch?v=Ghb7dOq8sTQ

It's from 10-mila. Night leg, spectator contol. A big train comes to control and they puch one by one. You can see NTNUI runner coming and punching just like the other runners. But for some reason he does not get signal, but there is an other runner punching with the other unit and there is beep from there and NTNUI runner continues without signal. DQ. No signal, no punch -> DQ. Did he do it in purpose? Did he try to get some advantage there? No. I don't say there is necessarily something wrong in disqualifying the team here, but I say things does not have to be like this. The disqualifying of the NTHI is not the problem, the problem is things have come to the point we need to do it. The same thing has happenned in Jukola with Emit as most AP readers know only too well. So we may have a problem here.

(end of part two)

You may think we can fix this by using louder beep, or even better by putting some kind of vibrator to each stick/brick so you could feel you have punched. But it still would not solve the problem. Getting disqualified isn't the only fairness issue here. This true story by Pasi Ikonen is a good example: PWT race 2002, Vierumäki Finland with SportIdent. #13 he got no signal. He tried again, still no signal. The he continued, but soon turned back to try again and try to do it with time. He kept the stick in for several seconds and because he still got no signal he continued to finish. He got first disqualified, but organizers checked the punch unit and yes, the much was there. Of course, his stick was in the unit for ten seconds altogether. He go qualified and in final results third, only 3.4 sec behind the winner Omelchenko. You can find split times from www.pwt.org and you can see any other runners didn't have any problems there. So, Pasi clearly lost the win there, and it is obvious it really wasn't fair. Would any louder beep or a vibrator stick solved this? No, not really.

You could say using Emit would solve this type of problems. But it isn't true. There is lcd display in new Emit bricks, and because punching is so difficult, most runners just take the brick close to the punch unit and check e-punching really happens by watching the display. If unit is dead, you need to either take pin mark with back up paper of try again with on other unit. If it's the last control in mass start race or relay, this could mean all winning changes are gone. So using Emit instead of SI doesn't solve anything.

You may also say, it part of this sport and there is some other aspects in orienteering that make bigger differences. But punching wasn't any unfairness factor when pin punching was used, so why we need to accept it now just like that? Can't we question it, could we just figure out a way to make this part of the sport fair?

If we consider orienteering as a respectable, seriously taken sports, can we really use current punching systems? And use tenth of second timing with it? And hope someone takes this seriously, not as a treasure hunt game?


(end of part 3)


So it looks like things are not perfect now. But do really have the technology to do it in a better way? Lets think what is needed. I'll try to describe basics what could be done differently and some technology examples. I am not expert on tech matters, so technology part is just an example.

First we need to decide what are the responsibilities and what punching operation is for an orienteer. Organizers put punch units to forest and competitors can see only example one. It would be wrong if competitors are responsible those units are ok. So it is organizers. On the other hand organizers can not know what competitor does with the stick after start. So competitors should be responsible for keeping the stick OK. Easy so far. Then punching. We know now competitor may loose time if there is a dead punch unit and he/she has to try punching with an other unit if there is no signal. If organizers are responsible for the punch units, competitor should not need to do that. The only way to get around this problem is getting rid of all signals. If unit is dead, competitor should not know about it. It should be organizers problem and dead unit should lead to qualifying all runners who visited that by control by the time it was dead. Competitors can not know there is dead units, so they can not take advantage of this.

We need to describe what punching is. If there is no any signals, it can not be waiting "until you get signal". It should be a simple mechanical operation, so easy to do you can not win anything by making it badly. SI style putting stick in the hole without bothering to wait any signals would be just fine. It also could be touching punch unit with the stick/brick. Something as simple as that. If this could be done, it would solve the couple of seconds unfairness issue for dead unit, but not the DQ issue.

The biggest reason behind DQ issues is the fact we have on one bit for our punch. We either have a punch or not. Zero or one, white or black. Let's compare how thing were with pin punching. You could make perfectly clear punches to the middle of the punch box. Or do it reasonably fast, when there may have been some single pin holes outside the box. And some tried their luck and punched where ever and the paper was full of random pin holes. If you punched well, you got qualified. If you didn't do as carefully as should and your punches were readable bit not quite perfectly in their boxes, organizer usually told the competitor next time they will DQ you, and they will tell to next event organizers thy should watch for your punches. And if you tried your luck and punched all over your card organizers never even tried to find punches, it was DQ immediately.

Could we learns something from this. Maybe e-puch system should not be that black and white, there could be grey area in between and organizers could decide where they put the DQ line (not computer as it is now). Example. What if we could have a parallel system in our stick just for measuring and recording a signal from the punch unit. It tries not at all to interpret what is the code number, it just saves how strong he signal is and how long time the stick was close to the unit. If punching fails like what happened to NTNUI runner, we could see from this data the stick was inside the hole for 0.6 seconds. And if e-punch takes for example 0.1 sec and in 99.99 % of cases under 0.15 sec, we could decide if someone manages to keep stick in for over 0.15 sec without getting the actual punch, he/she should be qualified.

(end of part 4)


Now, let's think a bit what kind of technology could be used, just as an example. The technology choices of course should be made by system manufacturers. These are just as an example, to get some kind of idea what it would take.

Lets think about current SI. What if we have a simple magnet under the punch unit hole. There could be something like three strengths and we could put the magnet north or south pointing up. So we have six alternatives. The stick could have some kind of simple mechanism, it wakes up if magnetic field changes and it records how big is the change and the field direction. Measuring and recording technology for recording these can not be very expensive or difficult/extraordinary(?). Now, if normal punch fails, we could afterwards check is there record of this magnetic field change and how big it is, how long it took and what was the direction and when it happened. We could compare it to the record made with other competitors sticks and check what would be the split times like. And then decide should the competitor be disqualified or not. For trainings and permanent controls we could use magnets only. Of could no-one could be sure was the control the right one or not, but you could get nice split times. And punch units would be cheap, just one simple magnet, and we could detach it to control flags. And they would not need any batteries - good for permanent controls and good for competition back up use - magnet would never be dead.

With Emit we could so the same, and also get rid of the back up paper and go from the current brick style to stick like - like SI but with Emit technology inside. And with this similar magnet back up thing.

Maybe we could even make these punch units totally with magnets. Something like SI style hole. Just ten magnets around the hole, north or south pointing to the hole centre. The stick reads the north-south sequence of the magnets. 2 ^ 10 = 1024 alternatives. And here we could use similar back up recorder described above, but use these ten magnets.

Yet an other example. Instead of expensive technology SI and Emit are using in their punch units, we could use for example RFID chips. And competitors brick/stick could be a RFID reader. These chips are pretty cheap these days. For trainings/permanent controls we could have controls with these simple RFID chip stickers. And for competitions these could be back upped with the magnet thing.

One possibility is bar codes. If we could have a some kind of bar code reader or digital camera included in our sticks, punch units would be cheap. We could make them with any b/w laser printer.

It's true, equipment manufacturers need to get their share, so if punch units are cheap, we could expect sticks might be more expensive than before. But I don't think it's that bad, we don't need to leave them to the forest as we need to do with out punch units.

But like I said, manufacturers should know these technology issues much better and they know how these things. I know nothing about this technology and electronics and I know these examples are ridiculous. But the technology is not the point here. The more important thing is what these elements are used for and how. Fairness issue, back up needs, and the e-punch units for trainings/permanent controls.

[here chapters about eliminating possibilities to make human errors, meaning simple punch units without any need for programming and smart sticks, unlike current SI etc...]


Can something like that really can be done, I am not sure. But I think and hope there is. Current systems are based on designs made early 90'. A lot has happened in IT sector during last 15 years. Also we have now 15 years of experience about these things, back then they had none. It's quite clear it could be done now in a lot better way.

Anyway, improvement like this will never happen if we don't admit we really need it and also tell it aloud. SI and Emit with new products like fast sticks, fancy lcd displays and touch free systems - it clear sign they are not seriously trying to do any core improvement. It's the same 90' design with same flaws. They will not make it better if no-one tells then what are the needs. But we can not do it either. What we can do is let IOF people know we are very worried about the direction of the current development. And ask then to gather some wise people to think this trough, for example figure out a list of current problems and the essential needs for a good e-punch system, together with current manufactures. And then they could agree some kind of time schedule when only systems fulfilling these new standards can be used for IOF events. Like after 201X only certified systems or pin punching is allowed in IOF events. I don't see why everyone could be happy with this, those companies could sell a lot new toys, we orienteers would get new more reliable and fair e-punch system and be happily pay for it. And current old systems could of course be used in low level events until they are done.

This is how I see these things at the momenet.

(end of part 5)

"Then punching. We know now competitor may loose time if there is a dead punch unit and he/she has to try punching with an other unit if there is no signal. If organizers are responsible for the punch units, competitor should not need to do that. The only way to get around this problem is getting rid of all signals. If unit is dead, competitor should not know about it. It should be organizers problem and dead unit should lead to qualifying all runners who visited that by control by the time it was dead. Competitors can not know there is dead units, so they can not take advantage of this"

This was exactly the reasoning behind the design of the Emit system. It was first developed by a Norwegian company called Regnly, together with the Norwegian and Swedish orienteering federations. One of the key design goals was that the punching procedure always had to be the same, regardless of wether the hardware was actually working or not. The decision to have no signals to confirm punching was a deliberate and careful design and it makes perfect sense. The runner always punch the same way, doesn't have to wait for any confirmation, the paper backup records a needle mark, which can be used after the runner has finished the race, if it turns out a unit was not working properly. (The other obvious weakness with confirmation signal is that the organizer have to hang a traditional needle punch for backup, which is ridiculous, the whole point of an electronic system is to replace the needle punches in the first place... somehow SI managed to get around this issue by lobbying federations to allow multiple units at each control site as a backup, but this is not always possible for smaller clubs that can't afford to buy hundreds of units).

However, unfortunately most orienteers never understood these deeper issues. When SI came out with their system many years later, orienteers loved the beeping and blinking units that gave you the (false) comfort of confirmation. SI started to dominate the orienteering community (outside of Norway and Finland).

The other thing that people tend to forget is that both Emit and SI were not originally designed to actually time the race. They were just intended to replace needle punching and the huge labour of checking the needle patterns (especially at events like Oringen). When I started to run my first event with electronic punching, the timing itself was still done in the traditional way. The clocks in the units are cheap and drift, they were just intended to give splits, and it does not matter if splits are a few seconds off. The systems can not be used for timing at higher level championship events (at least this used to be true, maybe the very latest versions of the units are better?). This is why at WOC, for example, the timing of the race is done with a separate system.

I'm not really sure what I want to say with this. It does not relate directly to the above posting. I guess I just find this whole history interesting. I do agree with the above conclusions that it should be possible to engineer some better system today with 15 years of experience.

RFID would make for very cheap control boxes. I have heard that ARDF clubs have been making RFID systems. Am I misinformed?

A couple of points:
1) I haven't observed any of these 'problems' happening at WOCs or WMOCs. The reason is probably the extent of additional checking at WOC to make sure nothing goes wrong. So event organisers using these systems need to invest more time in preparation.
2) Read the article by Christer Norman in IOF's O-Zine 02/05 (pdf from publications page on IOF web site).
Basically SI and Emit are both 10-15 years old technology, and they have developed as commercial applications. Along with Mr Norman, I'd like to see the next generation punching technology developed on a non-commercial basis, by orienteers, for orienteers.

E-punching isn't as common where I live as it is other places, so I don't have a good feel for this, but I have to wonder whether the problems cited above are actually very rare. It's not surprising that out of the zillion times these systems have been used that there are a few hiccups. Are the examples listed the only significant times things have gone wrong, or are the representative of a more widespread failure? In the times that I have used epunching, there has been only one (maybe two) times when the epunch equipment failed.

Cheap, light, zero maintenance control equipment, such as barcodes, would eliminate my biggest objection to epunching, that it puts more work under the responsibility of the course setting crew. That said, I don't have much hope for barcodes (or any optical system) because of one thing: mud. No matter how you create an optical system, mud will eventually get onto something that needs to have light pass through it. Yes, it could be cleaned off, but people will find that annoying at best.

I may be wrong, but I thought both e-punching systems in principle already were using (early variations of) RFID technology?
Whether they do or not I suppose that rapid changes in technology may have made the original 10-15 years old solutions outdated.

Its a classical problem with any computer embedded system where heterogeneous redundancy will solve for abusive use.

From design perspective will require a group of 4-5 highly qualified engineers who are familiar with orienteering sport, $2mln budget, approximately two years before all bugs will be flushed out and system operate at acceptable level. You would also need a company that would be willing to provide service and development thereafter. Commercially it is not profitable.

The two orienteers in my company looked at it. The company has the required expertise. Sergey's numbers are correct and the project would not be profitable.

One thing that annoys me about E-Punch is that the user has no record that they can evaluate for themselves. With pin cards you can quickly glance down to see that every box is properly filled. You can also use the card to keep track of your progress. Now you have to wait until you're back at the download station.

Self checking is good for both sides. When an orienteer was sloppy with pin punch, they could be held accountable because they knew it.

It looks like the v3 EMIT with digital display tries to address this for the competitor, but even then the feedback is lower quality than pin cards, and there is a fairness issue that some competitors get better feedback than others. And this isn't even an option for SI.

I agree that SI and Emit are not ideal, for the reasons mentioned and more. (But I'll note that pin punches were never that reliable or user-friendly either. Pins got dull or even fell off, strings wrapped themselves around everything, the plastic broke, cards tore, and in the cold, hands struggled to punch. Stands helped, but needed to survive "heavy punchers", and so were bloody hefty. And muddy smelly cards full of holes were never a pleasantry when organizing. The good old days were horrible.)

Systems built for other sports, like running, might be reliable....but probably not designed for the forest.

The best bet might be to push for continued improvement of the existing systems. At least we have two competitors (though unfortunately incompatible).

I'm a bit bothered that people don't complain louder about problems when they come up. For instance, it took a long time to force SI to admit that there were problems that were later identified as cell phone interference. People were awfully meek until then. If we don't stay on their case, the manufacturers won't focus on what problems there are. That's I think the main lesson.

SI makes money on selling e-punches - chip by itself costs less than $1 plus plastic parts. But controls are quite expensive to produce, current price level barely covers manufacturing expenses.

Idea to use passive RFID or Wireless tags is nice, than controls will be very chip, but expect price of the stick above +100$ - plus battery inside, some moving parts (buttons), and etc. dont forget water resistance - will be very critical.

There was another discussion about SportIdent errors after a DQ for "mispunching" at last year's WOC. It starts here, with distracted's post (8 from the top).

I'd like to see some statistical analysis of e-punch failure rates. Any new system ought to have better reliablility, at least on paper, for anyone to consider spending money developing it.

What I'm most upset about isn't the failures of the system, it's the unfairness of disqualifying someone when there is clear proof they were actually at that control.

Putting passive RFID tags at the controls with readers in the runners hands wouldn't be too far removed from a Sportident race with dibbers at the controls and the control boxes with the runners.

technology is progressing - one of possible solutions (not the only one but reviewed by me) is small enough to fit into automobile remote control size box. And this would be active part. Passive part (control) would be about the same size.

Plus its wireless and allows multiple punches at the same time.

cedarcreek I wouldn't be too sure about Oli Johnson's mispunch at WOC. Looking at the maps there were two very similar gullies close together, and he probably punched the wrong one. How many times have you heard people swear they were at the right control? Personally I know of hundreds - and I've done it myself!

I agree that if someone's PUNCHING of a control (not just being there) can be corroborated, then they shouldn't be disq. I don't know if they have it at WOC*, but in Australia there is a pin punch on every SI control stand, and if you don't hear the beep, you punch a box on your map.

People tend to get uptight about how much time they can lose by slow punching. Compared to time loss by mistakes or route choice, its insignificant. Take the time to do it properly, like every other aspect of your race.

* They used to have every control at WOC manned, and the officials recorded all the competitor numbers, so if they still do that, I imagine they checked the situation with Oli Johnson, assuming he queried his mispunch.

The issue with SI seems to be that due to a hardware or firmware bug, a situation is possible in which there is a beep and a flash, and a punch recorded in the unit but not in the fingerstick. It seems to occur regularly and SI swears that it cannot occur. The IOF takes SI's word on that.

I think we just had just that happen at a WRE last week. The competitor in question was seen punching, so was reinstated.

The situation I'm personally aware of was an error where the unit didn't flash or beep. When I did the download, there were 9 missing punches. I downloaded all nine stations, and each was an "ErrB", which SI claims means the stick was removed too soon. The runner reported *repeatedly* sticking the stick into the control.

I have to disagree with Tundra/Desert. For me the issue isn't only for errors with a beep and flash. No matter how much I try to remember, there are always some punches where I'm in a big hurry and I don't check. It's only later I realize I didn't remember to check for them.

The reports I've seen from IOF events says if the stick fails to report the punch, it didn't happen. I say if there is proof the runner was there, it's unfair to DQ the runner.

Very interesting points already here. It is quite obvious making new system from the beginnig is not profitable. This means is there will be any improvemens in the future, it will be made by current compenies, SI, Emit, and possibly by the Russian one maybe or these Chinese ones:

http://www.chinahealth.cn/english/
http://www.orienteering.cn/orienteering/product/

This also means there has to be a lot of unrealism in Mr Norman's article. I don't think he ment the system should be developed by orienteers non-commercial basis. I guess he ment we should only discuss and agree about the most important needs and reguirements. I am not sure have we done that yet.

I thought the further complication with the WOC mispunch was that the radio control sent back the information that he was there, but nothing registered on his SI card. This may be completely off the wall, but there were definitely rumours circulating to that effect. Also, can't you tell if you punched a wrong control cause it shows up on the bottom of your printout?

At JWOC there were two people (a Canadian and Lithuanian) who mispunched the spectator control during the sprint where they couldn't hear the beep over the announcer and the crowd. They were definitely in the right place, we had photos of them punching, but the officials said that they punched too fast and thus reinstating their times would be inappropriate since they didn't have to spend the same amount of time at the control. The same thing happened to the Norwegian girl who was winning the relay. The lead men where coming up to the same control, right on her heels, and again, she punched "too fast." So there are three examples I know of, with the latest version of SI units and cards, where something went wrong. (Yes, I know you should always listen for the beep and watch for the light)

At WMOC 2005, we had a master map at the results tent. When a competitor came to inquire about a DQ for punching the wrong control, it helped to show them where they had actually punched. That usually produced a moment of revelation...now it all makes sense...and a calmer acceptance of fate.

Of course, that still leaves the problem of competitors who swear they heard a beep, but there's no record on the card.

Vlad, did you try downloading the log from the control unit, to see if there was a record?

The ChinaHealth system, except for the card, almost looks like what I suspected might someday happen...a SportIdent clone. Since the chips and basic technology are all off the shelf I'm under the impression, it was always possible for someone else to package up the same components. (Of course, it could be a clone...the card might just be packaged differently. It'd be interesting if anyone had tried to see if they were compatible. But even if they're the same components, which is by no means certain, they might record data in a different format, or transmit data by a different protocol.) Does anyone know anything about it?

I was preoccupied with other tasks, but I asked for the unit in question to be queried. I do not know what the result was. The IOF Adviser said that the competitor should be reinstated regardless of the query because there was unquestioned evidence the competitor did in fact punch.

Not questioning the reinstatement, just curious about what the unit said.

It's unfortunate that we're stuck with bland assurances from the manufacturers about the systems. (That a certain control unit log entry implies that the competitor punched too fast, and not some glitch that prevented the punch sequence from proceeding. That a beep and a flash mean that a record was made on the card...or that the unit was starting up or shutting down...and not some diagnostic or error indication.)

Short of open source firmware and perhaps published electronics drawings, which is an unlikely scenario, perhaps the best we could hope for is having a trusted third party audit the firmware and electronics under a non-disclosure agreement? An idea anyway.

Yes, three boxes were read for Saturday at the BOK/WRE day referenced by Vladimir for competitors who said they visited the control. All were BSF 8 units and were about 12 months old.

Box 1 - punch was there with a time, not an error message - SI 6 card - this is the competitor Vladimir refers to with the witness
Box 2 - punch was there with an error message - SI 5 card
Box 3 - no punch evidence - SI 5 card

Yep. Seems pretty conclusive that there is a bug (the case of Box 1). It will most likely take a disqualification of a WOC medalist from a country with an IOF Council representative for them to fix this, anything less has already been tried... and all they (SI) keep saying is, punch slower...

SI suckssss...... too slow, too unpredictable, too quite, too unreliable, too faint, too random, too un-cool and TOO SWEDISH.

This could get interesting... I'd guess that most Americans have never used EMIT and would have the opposite reaction. I'll go get my popcorn.

But the error messages are still auf Deutsch. How (un)cool is that?

SI is developed in Germany by Germans. Not Swedish what so ever...

Yes, the Box 1 result is pretty explosive. I would definitely recommend reporting it to SI, and to IOF, Valerie.

My last event with Emit, I got DQed for a missing control. I didn't carry a paper backup because I hate the wrist-wrenching thing, and the units all had LEDs to indicate that you'd punched. Still missed one somehow, and one I was pretty sure had flashed. But oh well, I wasn't that worried, just a fun run for me really, and not absolutely certain.

If I used Emit much, I'd probably get a card that shows what controls I've punched (and still not use the paper backup). (The paper backup on Emit doesn't seem like that great a backup. Aside from the contortions needed, the card seems pretty easy to lose, and unlike pin-punching it's not so easy to figure out whether you've punched or not.)

On the plus side for Emit, it's easier for the organizers. (No programming or synching of the units. Just put them out like pin punches.)

So, in one American's view, Emit doesn't suck, but not utterly perfect either. (We should have a poll on each system...sucks/doesn't suck. Could be an interesting insight into the public experience.)

EMIT cards lock into the control. It may take a while to master a "quick punch" but its certainly quicker and once it locks in you KNOW you have registered.... the backup is next to useless.

No flat batteries
No charging problems
No turning on problems
No sound problems
No flash problems
NO WAITING

I think a few of you have bad experiences with EMIT due to using old EMIT systems, although they do have a longer life than SI.

Well, we've been using SI for quite a few years now and I must admit this problem with punching has occured more often than I would like it too.

We organised the EYOC a few years back in Salzburg, and during the sprint one of the competitors was DQ'ed because he didn't leave the stick in the control long enough for it to register. However the control did register it, and he was seen by ... well 100's ... of spectators at the final control. After reading out the control and seeing that he had punched, he was STILL DQ'ed, which I found to be incredible and grossly unfair, because he didn't leave the stick in long enough. Not a good situation.

On the other hand I've been DQ'ed using emit because I failed to show a punch and my back up paper was missing (I lost it in the forest). Also not a good situation.

I believe that the current rule is, if a punch doesn't show on the card, then the competitor is to be disqualified (even if it is correctly read out on the control). Which would mean that we are only using the one system as a check, although in most cases we do have the backup check.

Maybe we should take a page out of the book of Formula one. Here every system is duplicated to prevent breakdowns or errors occuring. Why couldn't we move our punching system in this direction.

1) Timing as normal unless a mistake comes up,
2) Then check the backup system, and if there then the competitor is re-instated.

or

1) Build two systems into one chip (and the same at each control station) electronically,
2) Have two systems ready for chip reading at the finish, if it doesn't work on the first, then check it on the second!

Obvioulsy this does create quite a bit more extra work, for both the system developers, and the organisers, but the redundancy might make this part of orienteering just a little bit fairer.....

About Emit:

- you can now get e-punch without getting back up mark. This is why back up paper is almost useless as mentioned here. People get used to punch just to get the e one only, like JimBaker, because that's is enough, and taking the the back up one is using time for nothing. I think it should be the other way round. You should get back up punch first and more easily than the real one. And then make rules for handling it.

Exmaple. If take back up pin mark as the main punch and consider electric punch as back up. Then we could have a rule for sloppy punching: You can have 3 sloppy punches (e-punch ok, but no pin mark) - if you have four missing pin marks you'll get DQ:d. This would guide people's punching habbit to be a bit more carefully, and what is more important, no-one could get any advantage by taking risks in punching. And you could use new LCD emits, because al you could see is you have back up punch. And if lcd user sees unit is dead, all he needs to do is just punch as just always, no need to find an other unit. And if broken unit is reported, you'll just qualify all visits at that coutrol at that time period, even if pin mark is missing (but if pin mark is missing it's a sloppy punch more).

I doesn't mean I think we should start using Emit like this, I just tried to describe why Emit's back up it isn't really working as it should. This change wouldn't make punching with Emit more confortable than before, of course. We could get better application if we could combine regular Emit and touch free emit. Touch free part would be the back up one and the regular on the main one.

Touch free would take punch if you take your brick 25 cm from the punch unit. Instead of tricy shaped regular punch unit you could have just 10 cm diameter flat, smooth, round surface and punching is made by touching it with the bottom of your brick. So you wouldn't need twist it on a difficult hole like we have
to do now.

Then we have punching rules, like the 3 sloppy punches allowed, meaning you have touch free punch but not the regular one. If either punch is found, competitor must be disqualified if punch units is not reported to be dead. And because it all electric, there is no need to check or count pin marks. But no any non-electronic evidence of the punch either if both systems fail, but this is the price of comfortable punching.

This would make punching with Emit at least as easy and comfortable as it is with SI (when it works). I guess old and touch free uses same technology/frequency and can not just used at the same time just like that. But they already have pretty good technology to make it work quite well. Note: in competitors point of view punching would still be a simple mechanical operation - "touch the surface with the bottom of your e-card".

One example would be using touch free emit as a back up for SI combined with this 3 sloppy punch rule. Emit brick band around wrist in same hand with SI stick.

---

A bit about Emit and permanent controls and using it training, especially training in urban envireoment. We could have a button in Emit to get split time. Then we could run club training courses with it without having to put any punch units to forest. We can take splits with a watch of course, but this way we could easily read all splits with Emit's brick reader and analyze the data with splitsbrowser/winsplits/Routegadget, publish or what ever, by using the same methods and sofware we normally do without having to manually type splits anywhere.

The button could also be placed in a special way: we could have 100% plastic "fake punch units" for training only. When we put the brick in the unit, the punch unit pushes the button. These punch units could be used in trainings, permanent controls there is no need to disqualify anyone and only splits are deeded for further analysis. We wouldn't have to leave electronics in our parks to be broken by vandals.

JimBaker, this is from EMIT's pages: "Optional light signal shows that the unit is working, and that an e-card is activated."

Activated? Does anybody know what it means? It may not mean at all the same as punch data saved succesfully.

Just some observations about SI design. It uses 125Khz RFID technology with passive storage (e-stick) powered by electromagnetic field from coil in the control unit. The time needed to load capacitor inside e-stick is main factor for delays.
Algorithm is simple - Detect e-stick->power up coil->load internal capacitor ->read e-stick-> analyze content->add control id and time stamp->write into e-stick. Details omitted.
Remove e-stick too early - and you have our situation - no data written. Interesting, if they perform verification after data is written into e-stick.

Another interesting discovery is that SI uses software protocol implementation, not standard hardware transponder chip to communicate with e-stick. Thus they have to run real time process for transponder emulation plus all other functionality in the background - not trivial task. but very critical in respect to general system stability.

On the other hand, catch such an intermittent failure is extremely difficult task, especially in the embedded systems - we spent weeks trying to fix problem with repoduction rate 1/2000.

The point is not to weed out the 99.9%th bug, the point is that IOF people have to understand that 99.9%th bugs exist. So far, they seem to be blissfully (and in some cases, aggressively) unaware of the realities of system development.

- Here is yet an other video of a too fast punch with SI, 25-manna 2004
http://www.lynx.fi/videot/manna2004.mpg

- WOC 2001 Finland, Women's relay, Lithuania got DQ for missing punch at TV control, but was seen punching (Emit) there (or this is what I have heard)

- Ski WOC 2007 couple of weeks ago, women's mass start race. Liisa Anttila punched too fast with SI, she was leading the race with about 15 sec and there was only about 5 contols left. Other competitors saw her punching there. She was first at finish but got DQ and we got an other winner.

The IOF did not really give the Ski-O WOC organizers any leeway in interpreting the Rules in that last case. (IOF 20.5)

You should get back up punch first and more easily than the real one.

The right principle I think.

One example would be using touch free emit as a back up for SI combined with this 3 sloppy punch rule. Emit brick band around wrist in same hand with SI stick.

Someone would need to investigate to make sure the systems don't interfere with each other in any way, but an interesting concept. The keeners all end up owning both cards anyway... :-)...at least in Europe.

from EMIT's pages: "Optional light signal shows that the unit is working, and that an e-card is activated."[...] It may not mean at all the same as punch data saved succesfully.

I started to wonder that after my posting. I'd be curious to know.

Hm... why would you disqualify someone seen at a control? It should be easy to estimate the time they were there and figure out whether it was visited in proper sequence. Is this a race-specific rule, or is it an IOF rule?

IOF 20.5 says:

A competitor with a control punch missing or unidentifiable shall not be placed unless it can be established with certainty that the punch missing or unidentifiable is not the competitor’s fault. In this exceptional circumstance, other evidence may be used to prove that the competitor visited the control, such as evidence from control officials or cameras or read-out from the control unit. In all other circumstances, such evidence is not acceptable and the competitor must be disqualified. In the case of SportIdent, this rule means that:

• If one unit is not working, a competitor must use the backup provided and will be disqualified if no punch is recorded

• If a competitor punches too fast and fails to receive the feedback signals, the card will not contain the punch and the competitor must be disqualified (even though the control unit may have recorded the competitor’s card number)

It has been viciously enforced at major IOF events. At the BOK WRE, we had a nice and sensible IOF Adviser who didn't want anything to do with it.

Surely all rules are meant to be enforced, so I don't know how you can describe them as being 'viciously enforced'. In the pre-history days, would it have been 'vicious' to disqualify someone at an important event for punching outside the box?
(from the offender's viewpoint maybe, but not to the other competitors who all took the time to punch correctly)

The rule seems clear to me. You punch an SI unit. You wait for the beep. If no beep, you use the backup. Same as pre-history - you punch the card, check that the punch has registered correctly, if not, punch again.

It doesn't seem possible to me that 99 competitors can get a beep and see the flash and be recorded, while one competitor hears a beep and isn't recorded. It is bound to be human error, not machine error.

If it happened to 10 out of the 100 (or even 2), then it becomes uncertain that it is the competitor's fault, and the 'exceptional circumstance' part of the rule can be applied, and other evidence used.

Orienteering is not (as far as I'm concerned) a sport that's about punching. It's about going places. If you successfully went to the places, then you completed the course. If there's clear evidence that you went to the places, then you should not be disqualified. The simplest clear evidence is a punch record, be it on paper or electronic. But if there is other clear evidence that you went to the specified place, then disqualification makes no sense.

In the pre-history days, would it have been 'vicious' to disqualify someone at an important event for punching outside the box?

Yes. Yvette Hague, Short Qualifier, WOC 1991, Czechoslovakia. Last (next to last?) control, punched in front of numerous spectators, finished first in her heat. Punch partially, but not entirely, inside the box. DSQ. Preposterous and scandalous.

It doesn't seem possible to me that 99 competitors can get a beep and see the flash and be recorded, while one competitor hears a beep and isn't recorded. It is bound to be human error, not machine error

Oooh... compare, for enlightenment, the above with the following statement:

It doesn't seem possible to me that 99 people sit in the office and run copies of Windo*s that happen to not crash their computer, while one person all of a sudden gets a blue screen... it mush be something he has done, not something M*** programmers didn't think of.

Real-time systems are complex. To get something 90% working is usually less than half of a project; the next 9% working, 80% of the hours; the next 0.9% on the way to perfection, 90% of the time, and so on. A reasonable project manager will know where to stop. An about 1E-5 bug rate is completely to be foreseen given SI's size and person-hours of R&D. Actually I'd say if they don't have that order of magnitude of a bug rate, then they must have wasted time on debugging that would have been better used designing new products.

Now given this, is it fair to penalize the runner?

jjcote, it would be 'preposterous and scandalous' if they had not dq'd Yvette. Who would ever have taken orienteering seriously as a sport if they had not?

Yvette knew the rules, and I'm sure she would have kicked herself rather than blame the officials, especially as it was only a qualifier and she had plenty of time.

Should a long jumper have their jump count if their toe is over the take-off line? Should a golfer not be penalised when they take an illegal drop or move their ball? Why have lines on tennis courts? Every sport has to have rules to prevent cheating or sloppy play.

In 100m sprinting it got so that runners would deliberately try to anticipate the gun (ie cheat). So they cracked down on it by tightening up on disqualifications. In orienteering, some people try to increase their punching speed, so a rule is needed to ensure that they still do it correctly.

The sport that I do is primarily about using a map to visit controls, but at the present time it includes punching at those controls. Until the rules change, I'll continue to punch with due care.

t would be 'preposterous and scandalous' if they had not dq'd Yvette. Who would ever have taken orienteering seriously as a sport if they had not?

I don't see it that way at all. If all other punches were nicely inside the box, it was just an human error and she didn't try to gain or even need any advatage there. So DQ was 'preposterous and scandalous'.

With pin punching decision should a competitor be DQ:d or not was made by humans. They could judge the sloppyness rate after the run and draw the SQ line to a certain level to make it fair. With SI these udgements are made by a computer in the forest during the race and it's not based on any fairness judgements

Should a long jumper have their jump count if their toe is over the take-off line? Should a golfer not be penalised when they take an illegal drop or move their ball? Why have lines on tennis courts? Every sport has to have rules to prevent cheating or sloppy play.

I don' see much analogy here. Long jumpers has 6 attempts. If jumper's toe is over the take-off line he/she can try again. If that happens every time, he/she wouldn't get result. Fair. But if she does it once, it will not mean he/she is DQd. Same in tennis. They hit the ball hundreds of times during a game and if you once hit the outside the line it will not mean DQ for the player. Of course there is also rules for cheating, there will be DQ if you use too big racket or hit your opponent with it. In 200m run if you step on the outside line you will not get automaticly DQ:d if you are not obstructing other runners. And it's all judged by humans.

In orienteering with SI, one sloppy punch judged by a buggy computer means DQ. And no-one can tell was who was sloppy, competitor, stick or punch unit - did the runner win something there.

In orienteering, some people try to increase their punching speed, so a rule is needed to ensure that they still do it correctly.

I agree. But shouldn't we able to figure out was it a sloppy punch or 99.9% bug? And we are humans and we make human mistakes. Shouldn't we be able to somehow measure the sloppynes and make DQ decision by that? Do we really need to DQ a half blind half deaf H75 runner who can not hear the beep or see the flash in a bright sunshine just because the unit he used was dead?

For example Emit with the "3 sloppy punches (missing pin marks) allowed" rule sounds much more fair rule for ensuring they still try to do it correctly enough.

In orienteering, some people try to increase their punching speed, so a rule is needed to ensure that they still do it correctly.

If this were the reason, there could be a time penalty (10s perhaps), rather than a DQ. The main reason is that IOF doesn't want to delay the awards while someone collects control units, I understand.

The first YouTube video shows one flaw with the argument anyway. One competitor takes 0.6s to punch, the other 0.3s to punch. The 0.6s one gets DQed.

A bigger problem with the argument is the US WRE. Someone punched, a punch with a time (not an error code) was recorded in the control unit, but nothing on the card. I'd like to hear an explanation from SI for this. Until then, this generates enough uncertainty in my mind that clearly the fair thing to do is to give the competitor the benefit of the doubt, not the equipment.

SI stuff has had bugs in the past; we know that. With one version of the firmware, check units beeped even if the card wasn't clear. And SI has given bland assurances before..."it works fine for everyone else; you must have made a mistake"...until enough evidence mounted that they had to admit a problem, and discovered the cell phone interference issue.

The question, in the meantime, in the face of disturbing evidence, is do we want to DQ competitors and assume it's their mistake, or allow alternate evidence and allow that there could just possibly be a bug?

Having been burnt once by SI (brush-offs rather than action on failing units), I'm not willing to DQ people on SI's word, which is what it amounts to. People are fallible, but so are electronics, software and the people who make them.

One competitor takes 0.6s to punch, the other 0.3s to punch. The 0.6s one gets DQed.

And tenth of second timing is used. Just take a look at last WOC: http://www.woc2006.dk/results_sprint_final.html
How seriously this kind of racing and results can be taken?

The alternative might be to require organisers to interrogate field boxes whenever someone disputes a mispunch. What are the implications of that?

I'm a sceptic by nature - and for that reason I'm secretly inclined to agree that SI may have the odd bug. However it also means I think the You-tube evidence is dubious. The Jukola is a relay, so the mispuncher could quite easily have been at the wrong control. Also, how did someone with a camera happen to be there, or if it is an official camera, how did the person who posted on You-tube get hold of the clip - I'm sure the organisers wouldn't have wanted it posted. I'd need to see absolute proof that the clip is genuine and that the runner punched the correct control.

I didn't cite long-jumping and tennis as direct comparisons with orienteering, but purely to illustrate that every sport must have rules, and that if you don't follow them you may get penalised.

I was also primarily responding to jjcote's suggestion that punching isn't important because orienteering is about going places, and if you successfully went to the places then you completed the course. I wonder then, if he would reinstate Troy in the WOC Sprint Final results. He went to all the places, so the fact that he took an illegal route is irrelevant??

In my experience some (not all) elites and juniors (in Australia - maybe elsewhere too) are ignorant of the rules, don't know much about mapping specifications and map symbols, and often don't bother to check control codes. They should blame only themselves (or possibly their coaches) if they fall foul of the rules.

Split second timing was only used in sprint races until 2007 revision of IOF Rules. I was aware that the last Men's Sprint Final was close - my copy of the WOC dvd arrived the other day and I was watching it last night! (Great to see Hanny's win again!) However, from a runner's point of view on the day, you can't know the result in advance, so if it were me, I'd still be careful with punching and take my chances on losing by less than a second.

Did I miss something? How would it matter whether a runner was punching at the right control or not? Even at a control not on the runner's course, the unit ought to record their punch and write the same information to their card as it does for the runners who are supposed to punch. It's not as though the epunch units know who is running what course.

There is no Jukola relay youtube links here. The first one was from 10-mila and it was broadcasted live on Swedish TV. I guess somone just recorded with VCR. All runners had that TV control, of course. The second video is from 25-manna, it's a home video taken by people I know and I have no reason to believe they just made it up. They tried to use it as proof immediately at the event site but it didn't help - of course - and they posted it to web afterwards, just because they had it.

Of course there must be rules. But rules can, and bad rules should be changed. This SI rule means you can be DQd even if you do everything right, it also means if punch unit is broken, you will lose couple of seconds compared to other competitors. With Emit such a silly special rule is not needed at all. I can't agree everything is as it should there.

There must be a workaround. How SI could be used in a fair way? Man at every control checking if the stick goes in the hole or not (and SI data used only for split analysis) - can be done in WOC finals, but how about the rest of the races? Mayby all other races than WOC are so insignificant such a fairnes is not needed.

I wonder then, if he would reinstate Troy in the WOC Sprint Final results. He went to all the places, so the fact that he took an illegal route is irrelevant??

Well that's a stretch. No, an taking illegal route does not constitute properly going to the specified places.

I'm assuming Troy went through an area marked on the map as prohibited. For that, I would disqualify him. But if an open area were prohibited, and he ran on the sidewalk bordering it, I would not disqualify him if he leaned over that area when rounding a corner.

...Man at every control checking if the stick goes in the hole or not..
- alternative solution to use active stick, which emulates e-stick, but has log capabilities inside. present log to officials - and relax.

To be honest, for world-level events, it's completely feasible to have a video camera at each control as a backup. If there's a missing punch, check the video record.

I'm a bit bemused that the IOF makes such a lofty point of DQing people for a pin mark out of a square, but rarely if ever DQs someone for blatently asking for help, even in a WOC Final, even when presented with the evidence. The high proclamations about the sanctity of rules would ring a bit truer if the IOF did. Do only some rules matter?

(Having witnessed this at WOC long ago, I've experienced it at an event recently, chased down to the point of harassment by an elite woman desperate to ask where she was. Can't these people orienteer? But I digress.)

Frankly, I think there is simply an unexcusably big A**hole factor at work in the IOF, far more concerned with enforcing themselves, than serving their current and future constituents.

I observed this during the course of WOC 93 and almost everything I hear lately indicates this spirit is quite alive on a whole range of issues.

On the current topic, I think the vast majority of orienteers, especially those who have put in years of hard training, want fewer of these damnably stupid, trivial DQ's, not more.

If you cannot see a world of difference between running through an OB area, and mispunching by a tenth of a second, there is no defense for that lack of judgement other than the right to free speech.

If the rules need changing, change the rules. If the enforcement needs changing, get new standards and enforcers.

One of the few constructive IOF highlights of the recent years was the long overdue improvement of the control description symbols. I am under the impression that the major contributors to this were Australian. No coincidence to me that this was not a heartland IOF effort. If I am wrong on the nationality of this effort, I'll be glad to be corrected, but would add that this would still be the exception rather the rule.

I do not apologize for the name calling. I believe the IOF nees a serious slap in the face, more than more eloquent arguments. Their eyes, ears, and brains are simply not functioning on a whole range of issues. I am sure there are plenty of noble efforts at work within the IOF as well, but this organization seems in need of some serious reflection and house cleaning.

I am not convinced from the videos presented as evidence that most competitors are stopping to watch for that visual feedback from the SI box. I don't want to review this whole thread, so can anyone remind us all whether any of the cited instances are associated with a clear visual signal from the SI box?

I wonder if battery charge has anything to do with this. In the utube video the three preceding orienteers all punched the forward box that then failed to write to a stick. The successful puncher in .3 of a second used the other box. As a naive user of technology, I would be tempted to conclude that the forward unit could not cope with four visitors in quick succession. Maybe the issue of fairness is the difference in punch times between different boxes.

My conclusions on the limited evidence are:
Get the visual feedback, even if you have to wait.
Where there are multiple SI boxes, don't use the last used box if you are in a train.
Get rid of the audio feedback from SI boxes. It only leads to confusion when there are multiple boxes at a control.

On a selfish note, as someone who often sits behind the download station running the SI equipment, I don't look forward to a rule change that requires me to interrogate boxes whenever someone claims they punched but the equipment says otherwise. I'm sick of the job as it is..

can anyone remind us all whether any of the cited instances are associated with a clear visual signal from the SI box?

Well, the competitors usually say so (or say they heard the beep), but the event usually happens without witnesses (or with witnesses focused on their own race), and the videos posted above aren't close enough to show.

Getting caught-on-film evidence would probably require leaving video cameras recording several hours of up-close-and-personal major event punching. And what if it's only certain control units that have a problem? It's a classic problem of trying to track down an intermittent bug (if it exists)...lots of information is needed, but if the problem is sporadic, little is available, often fragmentary. Lots to investigate with wisps of info.)

(For example, maybe the bug isn't in that control unit, which properly writes the record, but in the next control unit, which overwrites the previous control's record for some reason. Or interference between nearby units.)

The best evidence to date* of a bug is the download of the US WRE box, showing a time for the competitor (and no error code), but nothing on the competitor's card.

I don't know if SI has an explanation for that. It merits one, I think.

(*AFAIK...anyone know of stronger evidence?)

I agree with turning off the sound, for the reason given...chance of overhearing another unit. (I make a point of looking for the flash at the finish unit, which is typically the only multi-punch site...maybe the last control too.) I do find Emit's flash too faint to be easily seen in daylight. (And the backup cards are susceptible to fluttering off.) SI's flash is probably adequate though, even with variations...haven't seen any units with too-dim LEDs lately.

the backup cards are susceptible to fluttering off

It really isn't big issue. You can tape it, use two sided tape under it (many Jukola teams do this, tape makes it a bit more thick, so yo'll get a bit better pin marks ). Or instead of those papers we could start using tape.

but because it useally is not needed, most people doesn't care much.

Get the visual feedback, even if you have to wait.
Get rid of the audio feedback from SI boxes. It only leads to confusion when there are multiple boxes at a control.

AUUUGHHHH! NO!!!! I'm colorblind, and in daylight, I can't see the red LED! With no beep, I'm screwed!

> AUUUGHHHH! NO!!!! I'm colorblind, and in daylight, I can't
> see the red LED! With no beep, I'm screwed!
OK, the audio feedback is unreliable with more than one box at a control. The visual is bad for the colour blind. I have a catalogue here for an electronics store. It has reaction time games that give the loser a shock. How about a massive voltage jolt to indicate a successful punch. Don't leave the control until you feel pain.

It's been suggested before. This whole discussion about SI problems seems to have taken place back in 1999 on the "O-net" discussion group.

A few stats from some of the events I've been involved in over the last year:

NB 'successful' means correct punch or mp not due to the supposed bug.

2006 Aust Champs Carnival (10 events) approx. 27,000 successful SI punches
2006 WMOC (4 events) approx. 200,000 successful SI punches
2006 Australian 3 days (4 events) approx. 40,000 successful SI punches.

I reckon just these 3 events represent about 0.5% of the total SI punches during 2006. How many of 50 million estimated punches for the year were mp due to the bug?

Well, stats can be used and read in many ways.

Simmo, have you any data about how may unsuccessfull punch attempts there has there been? Meaning dead SI units causing need for punch with an other unit or pin punch.

In races we (me and my family) ran with SI 2006-2007 the procent is about 5%, about every 20th unit was dead (our hilarious 25-manna problems are not included here).

If one loses 2 sec every time punch unit is dead, with this procentage and 50 million punces it means about 1400 hours of time wasted for searching a unit that actually works. And this during year 2006 only.

Maybe we were just unlucky, and better guess would be about 200 hours. But it still is more than a week.

With Emit this same time is zero seconds. Don't you think we entitled think there might be something to fix here?

A few years ago we were having similar failure rates. In almost all cases it could be blamed on the battery capacity algorithm being at variance with the battery capacity. ie flat batteries. The algorithm was poor because it required the user to know the capacity the battery and input that at battery change. It took no account of the variability in battery capacity, or the effects of low temperatures. The latter problem was most apparent at an Australian Relay Championships when an unexpected frost the morning of the event transformed a high proportion of working batteries into dead batteries. I think that event ended up working on an assumption of honesty. The batteries in the SI boxes also needed changing too often. Since the conversion of our boxes, the failure rate has plummetted. Anyone still using the old boxes will inevitably have the old problems.

Good point. Using only new units in championship level races would fix a lot. And and avoid using old units, especially in cold conditions. And if you use them, use one single battery brand/type and test batteries with your units. We can use our fridgerator for getting some cold conditions.

div mentoned "active stick". Is there such a devices already and what are those logging features like? Or is here the $2mln budget problem, and SI does not see it profitable and instead keeps saying there is no bug and everything works just perfect.

To be honest I must say I agree with simmo, these bug dq cases must be very very rare and most cases are just sloppy ones. The difficult thing here is the fact there is no way to know if it has happenned, or was it just a sloppy punch. And I still don't agree we should alloy waiting for a beep as a part of a punching. It should be stick in & stick out, thats it - no waiting. If techology is not ready for this, then we should select better technology. With emit this technlogy part has been fine already for more than ten years.

Pin punching had failure rates that were probably higher. (Pins that weren't sharp enough, or fell off, etc.) But a difference is that the orienteer could see whether a pin-punch had occurred, unlike SI and unlike Emit cards without displays. (Also, in terms of this record, once you've pin-punched a plastic control card, the mark is fairly irreversible. With SI, punches can be deleted or perhaps overwritten. As has happened with clear units left near controls or the finish, or units accidentally programmed as clear. Of course, in theory the units serve as a backup record, if they're read.)

The question was raised, what would the ideal system be? Perhaps:

- a display showing the contents of the card (as some Emits do I understand)
- write-once memory (maybe 100KB for 1000 races?)
- consistency of punching speed
- very fast punching (perhaps enough energy on the card for the whole race, and then some)
- easy punching
- perhaps some inobtrusive manual backup system, for us sceptics, nipping a corner off a plastic backup card and stamping it with a number and maybe time, like old library card machines...the audible feedback could be ca-chunk instead of beep :-)

Admittedly, Emit comes the closest. (If it weren't for that nasty wrist-wrenching, and the need for a tube of glue to affix the backup paper.)

Maybe, to solve wrist wrench, Emit punches should always be mounted vertically, fat end down. The wrist would still need to turn, but on an axis that it turns more easily (like turning a doorknob). The card would be strapped to the inside of your middle and ring fingers. The motion would be:
- reach down to the top of the post
- twist the wrist (like screwing or unscrewing a light bulb) so that the card is on the flat part of the Emit unit
- squeeze the card against the Emit unit using your thumb and fingers.

This would have the advantage of allowing use of a thin, lightweight post. Keep the post deliberately flexible so that there's no way that someone can punch without squeezing, so that people don't knock over the post.

My wrist and forearm can only turn about 180 degrees on the royal wave axis, so I might have to run a quarter of the way around the post, to be able to reach. Or, the unit's mount would need to spin..not hard I suspect...just a mounting hole that fits over a thin cylindrical top of the post. But using Emit, I don't find that I mind running a little of the way around the control, because rarely am I leaving a control in exactly the same direction that I approached it. It's just that the current mounting requires exactly one direction of approach, meaning some combination of running way around, or running around the right side when I wanted to run around the left side (due to obstacles or another orienteer or my direction of exit) and/or uncomfortable flexibility exercises. In practice, that's very awkward. (I don't think that SI would have been developed if Emit weren't so awkward, as normally mounted.)

So maybe what we need is Emit, mounted better, using the cards that have displays.

I will have my first experience of the Navlight system tomorrow. Maybe I'll have to report back here.

Navlight , I guess it's this one:
http://www.rogaine.org.nz/main_files/Resources/Nav...

Based on my extensive experience, I have not found that pin-punching has ever had a problem with pins that were not sharp enough, or that fell off. Pin-punching has its drawbacks, but I don't think those are among them. Those are also irreversible issues -- if somehow pins did fall off, they wouldn't reappear for the next runner. Or in the case of dull pins, it wouldn't be a sudden failure, you can take dull ones out of circulation. Pins falling off is probably no more common an issue than an epunch box being stolen.

For Emit, it seems reasonable to mount the box flat, on a pivot. I remember a meet years ago with pin punches mounted on top of a control stake, and I had the same issue with having to run around the control to punch. After the race, a friend of mine said that he just grabbed the punch and twisted it (the stake rotated in the ground) to get it to face him.

JimBaker, I agree there could be display, but only if back up is bullet proof. I mean if you can see from the lcd display a punch unit is dead, you shouldn't need to punch again with an other unit.

Emit punch unit maybe could also be a box with a 2 x 5 cm hole, you push your card head first in and when it reachs the bottom the unit hits a normal pin mark to the back up paper like a typewriter. You might even be able to push it in the hole in two ways (also 180 deg rotaed) if the mechanish could figure out which side is the one with back up paper. I guess this hole wide enough at the top and narrows on the way down it might be quite easy and simple to use. But I have no idea how difficult it is to design such a mechanism and make it robust enough. It wouldn't be fun if your card stucks inside the unit (actually that happenned to us once with SI).

on my extensive experience, I have not found that pin-punching has ever had a problem with pins that were not sharp enough, or that fell off.

The newer ones are better, but I've definitely met toothless punches, and the main reason for checking the card used to be to make sure you'd punched hard enough. But the point was not so much reliability rates, but that you could see on your card the record that you had punched (unlike SI and most Emits).

A horizontal swiveling Emit would be easier than currently, but still a bit awkward I think, if done one handed...grasp from above, swivel, move card onto unit and press...or if done two-handed.

I still think that vertically would be easier. The hand, wrist and forearm are designed to twist on that axis, and to grasp things between the fingers and thumb. A much more natural action than currently.

The original advantage of horizontal punch mounting was to be able to push firmly enough on the pin punch. Not so firm a touch is needed with Emit, and its design makes one handededly squeezing the pins onto the card easier than with pin punch. (No need to insert your card into a gap (because the pin(s) point out, not in). Instead, you can position and squeeze in one motion.)

Card is already in right position when hanging douwd from the finger. If the hole is designed well, the hole might be able to make most of the card rotating work, and part of the punch unit could rotate inside the box when card is pushed in. Dead angle could maybe be under 90 degrees, which means we wouldn't ever need to rotate card more 45 degrees. And with two holes we might reach full 360 deg. But I am again only guessing here, there may be a lot of reasons why this can't or shouldn't be done.

Navlight looks interesting, it's a lot like jeffw wrote: "dibbers at the controls and the control boxes with the runners". But I will have same kind of no backup/sloppy punch/dead control(but tags will not die as easily) problems. But anyway, to me it sounds like an improved version of SI.

Using moving parts is bad idea, gets broken too easily. Let's forget that idea.

JimBaker, I can see your point. The disadvantage is we will not see the card & punch unit so well, it might make the positioning work more difficult. Also if unit is two sided (would make rotating work easier), you might hurt you thumb with the pin whan you squeese.

what do you think of this solution:
http://www.routegadget.net/images/emit1.jpg
http://www.routegadget.net/images/emit2.jpg
(this is how blurry my eyes are when I try to punch...)

Like this we could use our thumb, all fingers and wrist for twisting in a way like you suggested. Then we also chould re-design punch units in a way we could put card in both ways. This means we'd have less back up alternatives, but there would be still 70 alternatives left and that should be enough. Then just a little bit higher wings to every corner to make it more easy to push the card down right.

active stick is just "idea fix". dont think such real thing exist or existed. maybe in the lab which developed RFID staff. but it would be curiosity to develop this device.

My first experience of navlight.
It won't compete with SI or Emit.
Main advantage for rogainers was not having to program units to turn on. They could be placed a week in advance. Now that SI has solved that shortcoming, I'm not sure what the market is for Navlight.

I have organised a rogaine, the 7WRC, utilising Nav Light as Pensioner has written the advantage is that on a 220 squ kilometre course 450 kilometres from home you can hang flags early.

Though in the case of 7WRC it was the days, not weeks before the event.

Peter Squires who is involved with New Zealand Rogaining is the developer of the product.

It has certainly been taken up in NZ and Australia for rogaining purposes.

I believe that Eastern European Rogaining associatoins-who are closely linked to orienteering if not been orienteering clubs- have modified SI and Emit units for rogaining purposes though I doubt if that deals with the specifics of the scoring/control visit record.

Hi Alan. I'm one of those who missed out on the 7th WRC, much to my disgust. Our contribution was to host route gadget for you.
Perhaps you can enlighten us all on how navlight works. I have only seen it from the competitors end.
1. Does the control make a record of the competitors punch, or is it just the competitors button?
2. Now that SI can be placed some time in advance (it does not need to be programmed to turn on, and it has long life batteries), what ar ethe advantages of navlight? Is it cost?
3. Have you been able to compare software?
4. In the 7th WRC, were there any problems with recording visits to controls?
I get the impression that issuing 'buttons' to all team members acts as a form of back-up that is not available in orienteering.

Hello Neil. Sorry about the missed entry-that is a seperate discussion I might start here about whether early bird entry fees, early registratoin or qualification is the best way to manage entry for an open entry event ! Keep an eye on the 2016 Australian Championships for Warrumbungles 2 unless we return for a NSW Champoinships!!

Thanks for hosting of the WRC routes.

It is worth saying that Nav Light is a control visit and scoring system with a limited associated results software.

I would be curious to know whether SI and Emit extend into "event/entry/membership" management.

Yes the tag and punch record a visit. We did not provide Intention boards at WRC-these are lists hung at controls to let teams leave a record of the next control they plan to visit- not always reliable as teams change their minds. We asked teams at 7WRC to compete flight plans before they started which also have the limits. Interestingly the strongest Europeans at 7WRC deviated least from their flight a demonstration of competence and skill, I diverge.

To the end of question one. If we had lost a team besides searching roads, water points etc in the first instance we intended to take a latop out onto the course and down load from punches the visits recorded after reference to flight plans and other records.

My view is that you could use PDA's or other industrial data devices to collect data i.e. control visits.

Question 2. Not sure. I'm not technically knowledgeable. I know a rogaining organiser and orienteering organiser who is frustrated by the limitations of the software not so much the hardware of SI. That is why I asked about the limitations of SI and Emit.

Question 3. Cannot answer directly. The WRC adminstrator had to manually produce results by downloading from Nav Lights software as it has limited capcity for classes and categories. We also had a software/hardware problem at the finish as the printer "choked" on three hundred print jobs to provide entrants with a print out of their control visits before finalising results.

Question 4. We issued only one tag per team plus the traditional punch card with punch at each flag. For future events we will be moving to at least two tags per team ( rogaine teams are 2 to 5) we were nervous about moving to Nav Light completely as we had only used it for a 12 hour event 6 weeks prior to 7WRC. We lost 10? teams records at the finsh to a soft ware problem associated with the printing issue. So the punch cards were essential backup.

Clearly issuing a tag/button to each team member will resolve a problem for rogaining of team spliting up as every body will have to go to the punch. NZ and the ACT at least in Australia are already doing so.

People clearly love the ability to see the record of their the course. Why some people love to purchase xepnsive personal navigation aids/watches/shoes and IPODS!

I think I have hijacked this thread. I'll stop now!

I ran in an Emit event today using an approach similar to Jagge's (but less sophisticated...I put it together as we rode to the event). I used my SI lanyard to tether the Emit card to my wrist, and held it in my right (punching) hand. This did indeed make it easier to twist the card to match the Emit unit's orientation, though I had to make sure I had the card the right way up (Jagge's harness probably solves this), and I found that it helped to see the Emit unit from afar in order to prepare. It took a while for the motion to become more natural (it's fiddly at first), but by halfway through the course, it had become a definite improvement over wrist wrench.

It's still more involved than SI punching...though the manual backup is nice, avoiding the perpetual "I'm sure I punched".

If I get a chance (my club uses SI not Emit), I'll play around with vertical mounting. I think it would be easier than what I did today, or even Jagge's harness, because the card could be strapped firmly against the fingers. I suspect that the visibility of the unit will be adequate. From today's experience, I think that 270 degrees of twist will be feasible...combined with slightly positioning onesself at the unit (something I always do a bit anyway), enough. But to do is to know.

Hi

I've been thinking about this a bit since using the Navlight system on the wekeend for a rogaine.It worked well enough but it seems the whole system is fairly expensive ($A30000 according to the Queensland Rogaining Association).

In my "day" job I work with MiFare smartcards & embedded hardware - I got to thinking it wouldn't be too hard to create something similar that offered a bit more functionality (for a rogaine!).

Two teams were overnighted and, from the organising tent, there was no way of knowing for sure the last control they had visited. So, I got to thinking, how about recording visits from previous teams onto the next teams tag - that way you would be able to know which control the team visited last. It would also provide a backup if a tag was damaged in the field.

I know people above said it would cost $2M & a couple of years but I'm not sure I believe it. The hardware is pretty simple - a micro, timeclock, buzzer, light & MiFare reader. The embedded sofware isn't too hard - write the current time into the tag. The PC software would, in my mind, be the more complicated bit - trying to work out categories, etc.

Some quick calculations put the cost of an assembled reader at less than $A100 and a wrist mounted tag (looks like a watch) at $A3 - which means 80 controls & 500 tags is less than $A10000.

I'm not thinking of using custom mouldings, etc - just a waterproof box from the local electronics shop. It won't look pretty but that's not the aim. Plus they are fairly strong boxes so you could attach a loop on it and chain it to a pole/tree for urban events.

I can do the embedded stuff ok - the PC stuff would take more time than I could devote to it. Is anyone else interested in talking about it further to see if it really is achievable?

regards
Peter

Hello Peter,

I pointed this discussoin out to the NSWRA webmaster/Navlight/Event management person, Graeme Cooper.

I'll point you post out to him as I know he has been playing with the software of Navligght though not the hardware.

Cheers,
Alan

Hi,
A new open source punching system is emerging, see http://sourceforge.net/projects/openpunch/
Estimated cost: $3 for control, $10 for card

I'd be suprised if they can do an assembled, tested & packaged control with a battery for $3! The micro itself is $US2 for 25 or so.
I like the MiFare idea more for lots of reasons - I'll have to check on prices on Monday for the control - but the big question in my mind is whether there is "room" for another solution - for "normal" orienteering in Australia SportIdent seems to have all the market and rogaining is covered by Navlight.

Peter

You are right, I exaggerated, the cost of control is $US4 - 5 with assumption that you assemble it yourself.
Prices from my nearby electronic shop:
micro: $2.5, battery: $1, crystal: $0.5, enclosure: $0.5, ir led: $0.2
I agree, it will be very hard to introduce a new solution unless it is very cheap, reliable and easy to service. I think there is no room for another "commercial" system, but is still for open source one.

Back in the olden days, before electronic punching there were still problems with the IOF Rules; as has been pointed out above - the Yvette example. Rules are not handed down from the mount on tablets of stone; they are the subject of much review and discussion and decision by experienced people. IOF had such a review in 2002-2003 to which I made a submission - because I found the old rule discouraged people from going orienteering. The IOF considered all submissions and came up with the new rule which is the subject of this post. I think the situation is now worse; we have clear evidence that electronic punching is not 100% accurate, as any sane person would have thought anyway. But the new rule leaves organisers/event advisers with less discretion than they had before.

The only way to effect change is to get IOF to change the rule. That is done by member associations making submissions to the IOF Tech committee. I suggest that if people are interested in doing something about it, that we agree here (except Simmo) on a better wording for the rule and try and get it through our member associations. It is not going to be easy.

For some background I will successively post my original submission (the relevant part) from 2002, IOF's response and my final word (final because IOF didn't reply).

Proposed Amendments to the “INTERNATIONAL ORIENTEERING FEDERATION (IOF)
FOOT ORIENTEERING EVENTS” 2002

Consider – what are rules for: answer to ensure fair competition

I would like to propose the following 7 amendments to the IOF Rules. These amendments are presented with a view to helping make Orienteering more open and more welcoming to new members, recognising that Orienteering struggling to win the battle for hearts and minds with many people who have other things to do on the weekend. The rules were looked at with a view to see which, if any, rules could act as a deterrent to participation by competitors, planners or organisers. The purpose of these proposed changes is to encourage more people to participate and stay participating. The IOF rules are used as a template for member Federations, and many of the changes relate to rules designed for elite competition, which are applied to casual events by organisers who simply follow the lead set here. For each proposed change the proposal is followed by (a) the reasoning and (b) the full rule in question. There are 5 rule changes (mostly deleting a phrase), and 2 rule deletions.

Delete phrase “and that the mark missing or unidentifiable is not the competitor’s fault” from Rule 20.7. Organisers and controllers should not be seeking to apportion blame. It is the fact of the matter which is important - did the competitor visit the control or not? The important point is that we should be trying to give people the benefit of the doubt and providing them with a positive experience. Most people do not cheat and some can get upset at pretty trivial setbacks. Surely the important point in this rule is “can be established with certainty that the competitor visited the control ”. If this is established where is the problem? The rule, as it stands, is contradictory and can only lead to unnecessary adverserial conflict.
“20.7 A competitor with a control mark missing or unidentifiable shall not be placed, unless it can be established with certainty that the competitor visited the control and that the mark missing or unidentifiable is not the competitor’s fault.”
In Rule 20.8 Delete sentence “One mistake per competitor is acceptable, eg. marking outside the correct box or jumping one box, provided all markings can be identified clearly.” And change the “must” in the first sentence to “should”. Identifiability is covered in Rule 20.7. The extra sentence introduces unnecessary complication into the process. For instance it is open to an organiser to claim that a competitor who has skipped one box (say no. 5) on a 10 control course has in fact made 6 mistakes (ie punched the incorrect control in boxes 6,7,8,9,10 and 11). Once again the point is that the fact of the matter is what is important: if the competitor put considerable effort in travelling to the event and going around the course. We should consider what we are doing here: we are using disqualification as a re-education tool. The problem is that potential orienteers may not see it that way and can perceive disqualification as demeaning, humiliating or symptomatic of petty organisation, none of which will encourage them to keep orienteering.

“20.8 When systems with visible punch marks are used, at least a part of the marking must be in the appropriate box for this control or in an empty reserve box. One mistake per competitor is acceptable, eg. marking outside the correct box or jumping one box, provided all markings can be identified clearly. A competitor who attempts to gain advantage by inaccurate marking may be disqualified.”

OK Here is the IOF Response:

-----Original Message-----
From: IOF
Sent: Tuesday, 11 March 2003 6:32 AM
To: Eoin Rothery
Subject: Foot-O Rules

Dear Eoin

The Final Draft of the new version of the Foot-O Rules has recently
been mailed out to federations for their final comments. The Final
Draft has had some modifications since the First Draft to take into
account comments received and further discussions. I'd like to thank
you for the comments you sent and I can assure you that they have been
considered by the Foot-O Commission, the Elite Events Commission and
the Rules Commission.

I thought I should give you some feedback on the points you raised:

2. Punch missing or unidentifiable. Ideally we would only be interested in establishing whether or not a competitor visited a control. Unfortunately we don't live in an ideal world! It is very important that results can be produced quickly and easily and it is also important that the rules are fair. Electronic punching brings many advantages but it also introduces some problems. Specifically, if we allow other evidence to be used, nobody will bother wasting half a second in punching if there are officials, spectators or cameras at a control. Certainly in a sprint finish, there would be no point in punching the last control. With the SportIdent system, it is very easy to punch too fast by putting the SI-card in and out as fast as possible. In order to punch properly, you need to dwell for a fraction of a second in the hole - not a difficult thing to do. However, even if you punch too fast, the SI station will have grabbed and recorded the SI card number. If we allow this 'too fast' punching, competitors would come to rely on it, and no results could ever be confirmed until all control boxes have been collected in and interrogated. For these reasons, we need the current rule, and in fact I have expanded it in the Final Draft to give the specific consequences of the rule in the case of SportIdent.

3. Mistakes with visible punch marks. We are unwilling to alter this
rule at this stage. It was produced after much discussion. Electronic punching is fast becoming the norm in all major competitions. Of course in minor competitions and those for beginners, no one minds if the organiser errs on the side of generosity.

Best wishes

Finally my comment on the IOF response from 2003.

From: Eoin Rothery
Sent: Wednesday, 12 March 2003 10:15 AM
To: IOF
Subject: RE: Foot-O Rules

Thanks

I appreciate the consideration you have put into my comments. I think the important message is the one you have said and I quote "Of course in minor competitions and those for beginners, no one minds if the
organiser errs on the side of generosity". I would add ".... or of fact". This is my main issue with the rules as framed. I don't really mind (that much) if tight rules are used to produce quick results for elite competition, but I see a style of orienteering organisation encouraged
by these rules for all events no matter how insignificant which encourages petty officialdom and is counter-productive to the goal of increasing participation (because many people will be put off by this
attitude).

Could IOF put in a rider to the Rules of the form "These Rules are designed for competitions overseen by IOF advisers, but should be considered by National Federations as a basis for general rules of competition. For competitions of lower rank, the organiser should
have a degree of discretion on issues of disqualification, with a view to valuing peoples efforts"

Thanks again

Eoin Rothery

I believe we have a greater problem with volunteer labour participation than with competition participation.
We need to balance the value of participants against the value of organisers. The people sitting behind the SI software on the day are not petty officials. They are participants who have often been dragooned into managing a software system they do not fully understand, and by which quite a few feel somewhat intimidated. If the implication of the change you are suggesting is that these 'officials' will be expected to retrieve SI boxes from the field and interrogate them before producing results, then you are making the job of finding volunteers more difficult. I think the rules is pretty simple in most cases, If you take too little time to get a reading in your stick, then you haven't punched. How many times do controllers go back into the field to show a disgruntled competitor that they did not visit the control they think they did, or that the control was actually where it was meant to be? Mostly this task demonstrates that the task of orienteering at speed has disrupted the competitors judgement. Do we add the SI box check to the list of tasks for the day?
Compromise is needed, and I think the current rule is a reasonable compromise. At a local event I am prepared to take someone else's word that the control was visited. Actually, we hardly ever check punch cards. But I am not interested in retrieving and interrogating boxes if I use SI. If competitors are prepared to accept the word of another competitor as fair, so am I. But if the word of a competitor is not enough for the level of event seriousness, then I'd prefer to rely on the SI stick rather than box interrogation. The alternative is for the competitor to say to the organiser I am prepared to be petty enough to demand an independent check of my fellow runners claim that he visited the control after an SI stick said otherwise, and I want the official to do it. If this is the argument, then it is unreasonable to then accuse officials of being petty because they then use the evidence of the SI stick alone.
That was convoluted. I suppose my point is that if you want to be encouraging to participation, then just run the event on trust and don't worry about the technology. Not accepting a claim of a punch until confirmed by a box interrogation is no less petty than a disqualification based on the SI stick interrogation. Our sport is petty.

In fact the current IOF rule could be invoked to do precisely what you say I am suggesting - it demands that an investigation is carried out as to whose "fault" the mispiunch is and the only way to do that is to retrieve the SI Box. So you are already in your own nightmare scenario through the IOF rule

Hey maybe if we ignore this problem it'll go away. Let's just believe in the machines? - or maybe not:

http://www.attackpoint.org/discussionthread.jsp/me...

Is there an Association willing to take the IOF on and get this rule, which IOF specifically introduced for WOC, changed?

We need to tilt orienteering back in favour of the participants, not the hardware or the rules.