As one of those involved in hosting the 2018 North Americans, I can advise we are aware of it, and are on top of it.
Article says it applies to IOF major events. These probably include WOC, JWOC, WMOC, World Cup etc. Possibly includes Regional Championships, and WREs, but may not.
IOF have just increased the potential number of WREs, I doubt the MC is going to go to the trouble of applying this process to every one of them.
You mean the two different sizes of dot knoll on the example map?
You're promoted to the MC, Graeme.
Doesn't apply at this stage to "ordinary" WREs.
It seems like a sensible move. One can debate what the map standard should be, but if there's to be a map standard, it makes sense to follow it for major events (and for other new ISOM maps). I wonder whether national federations will follow suit, requiring at least the running of a checker program for the maps of national events.
Blair, can you tell us anything about your thoughts, or Council's discussion on this?
Very good Graeme!
Unfortunately, I think you just eliminated yourself for consideration for MC, SEA, and SOFT (if you ever emigrate).
You used your eyes, right?
Besides the violations with dot knolls, and elongated dot knolls, I see illegal form lines, (by multiple parameters), and I'm rather sure that some of the stony ground falls within the critically prohibited 13-19% and 25-31% ranges.
Having just celebrated American Thanksgiving, we should be so grateful we now have software to catch the perpetrators.
If they better represent the actual terrain, what's wrong with using elongated "dot" knolls? How else would one map them?
GuyO - it's not the elongated dot knolls that are a problem - that's in the standard and I don't think anyone's complaining about it.
On the left side of the map there is a brown dot that is larger than the other brown dots on the map. That's not to standard and is therefore a problem based on the stated intention of the article.
I would add that there are some aspects of that map (the contours mostly) that, while they may be technically to standard aren't nearly as good as they could be. In today's world of lidar and gps, contours that are that rounded just don't show the terrain forms the way orienteers expect of maps for major events. I hope this isn't just about checking where something technically conforms to the standard (not that I'm against that!).
Sorry, Canadian, what? Contours are supposed to be smoothed (in general). It makes them more legible. Jerky contours are a major drawback of mappers using lidar.
Yes - that they are.
As an example of what I mean though is all of the hills that are clearly just drawn with the elliptical drawing tool in Ocad. I'm sure that most of those hills aren't that perfectly elliptical. They can still be smooth while showing the true shape of features.
"True shape" is never shown.
Unless they are particularly banana-shaped, that level of generalisation seems fine to me.
Granted this is just a fun diversion, but I agree with Canadian.
The contour shapes here are primitively rounded and smooth for this terrain type. This is unrelated to over/under processed lidar shapes.
This looks like intermediate level mapping from the '80's or 90's, or a continental mapper drawing their first Nordic map, well maybe second map, because the cliffs indeed fit the contours. ;-)
BUT, I give the mapper extra credit for using his/her own eyes to adjust symbol sizes, and probably drawing better stony ground than currently allowed.
Maybe the map sample was intended to demonstrate nonconformance that should be caught.
The level of contour detail may be perfectly correct (hard to tell without walking or orienteering there). Whatever the mapper depicts should be depicted everywhere. So, if the contour depicts tiny undulations, then similar undulations should be depicted even when they fall between contours. Without stacking form lines, and without drawing a continuous form line between every pastor of contours, that means jettisoning minor cruft. It's a flaw of many maps that the mapper says "aha,I can see the miniscule topography corresponding to this jitter in the contour, therefore I'll leave it on the map", then either ignoring similar fluctuations elsewhere or jamming the map full of form lines, thus reducing legibility while adding little of true use and disguising the more major topography amidst the palsy.
Here in Norway we have decided to use the Swedish/IOF process to check map conformance for our major national races, after a very good discussion we had on a national mapping conference during the Blodslitet weekend.
We agreed that the errors found this way should be evaluated, none of them would automatically disqualify a map, and we agreed on how to handle several border line cases:
I.e. an area of green stripes which is large enough in total according to the rules, but which contains a relatively narrow panhandle where the mapper decides to use solid green instead to make it more readable: This could make the remaining area too small but would still be considered OK.
Another example is a long cliff where the middle part is impassable but one or both ends are passable, here it is quite possible for either of the ends to be too short according to the minimum rules, but it can still provide the best way to depict the terrain.
It is worth noting that if the mapper extends the passable part into the impassable, the error would not be found by the ocad tool, the same way the first example could be handled by keeping the entire area striped but overlaying solid green on the narrow panhandle part, i.e. we would want both the mapper and checker to use "common sense" when handling a presumably short list of exceptions.
Another comment... the map in the linked article seems sufficiently detailed to navigate rapidly and quite precisely. I obviously can't tell whether there are missing 2.5m high or deep topographic features, but the comment was that it looks insufficiently detailed for Nordic terrain, which I take to mean that one could jam in endless Wiggler's depicting one meter high topography, or less. Orienteering used to be about navigating rapidly and precisely through forests, meadows, moorland, sandhills, etc., and lately urban environs. But then it's become something else, some microfiche deciphering puzzle. If more is being mapped than is needed for rapid precise navigation, not because it's large but, well, just because, then the sport changes. And it's costly, and probably puts off a number of people who might enjoy navigation as sport, but are disenchanted by the microfiche deciphering. If NAOC does follow ISOM reasonably closely (and I believe Ross), then it'll be interesting whether attendees see this as adequate (and maybe even ideal), or whether people hanker for the barely readable chock full and then some style.
Terje, that's great to hear, both that Norway will be setting the expectation that the standard is applied for major event maps, and also that there's been such sensible sounding discussion about it's application. It's heartening.
Re. the map sample: Yes, I agree that the contours look a lot like some I have seen from classic photogrammetry, or when the mapper is treating single-contour knolls as symbols instead of shape depictions.
At that height and length surely precise shape does not matter. It's a knoll; get over it!
Conformity to specification is not the same as being good. Like maps around Tampere...
Thanks Nixon. I think this is one of the points that should be emphasized.
The reasonable Norwegian approach as described by Terje seems to recognize this. However what I have heard (granted hearsay) about the "nth degree" Swedish enforcement does not, and the IOF language seems to revere and emulate the Swedish experience.
Conformity to the specification, in the manner being checked, means having good legibility. The fact that it remains to ensure the other aspects of being good isn't a reason not to take steps to get legibility right.
A bunch of US mapping connoisseurs making fun of quite reasonable quality -control measures adopted by Scandinavians. Totally appropriate bearing in mind that just 2 weeks ago US Classic Champs was held on maps straight from 1970s
That is an ugly example map, in so many ways.
I wasn't on the maps yurets alludes to, nor did I even look at them closely. However, I aver that the contours wouldn't have been any worse if they were derived from technology straight from the 1970s.
I hope the throwback cartography didn't cause yuret's finely tuned engine to misfire.
Agreed, the drafting specs related to separation of symbols, are clearly important for legibility, and in general the symbol sizes are as well, but not to the .10 or even 1.0% level, and I think we already agree that many of the specs are in the range of honest debate.
Do we really need to start more disagreements about many details, insignificantly related to legibility, when the primary determinant of legibility (I think we agree) is selection of detail, which I'm rather sure defies the specs and algorithms in question. This hinges on humans being reasonable with each other, at the beginning of projects, with experienced eyes in the terrain and on paper, not hiding behind arbitrary numbers, decrees, and position titles.
This is to say nothing about the other serious determinants of map quality, such as symbol selection, location/geometry, and shaping, which are also outside the scope of this review process. Yet we already have these wonderful sounding claims about ensuring map quality.
I'm happy to see something being done (at the IOF level, and in some cases at the national level), even if it isn't yet everything. I see it as a step in the right direction, which may even encourage following the spec in other ways, like minimum feature sizes. The discussion of why the mapper chose to jam in so much that minimum separations couldn't be kept at 1:15,000 might lead to a walk into the field to see features a small fraction of a meter mapped, or sub two meter topography, and a discussion on what features to select. This march toward the minute he been going on for decades, and some steps ought be taken.
The spec is actually a quite good description in writing of what an orienteering map should be, with some points always in valid discussion of improvement (glass far more than half full view). I'd be very glad to see it applied more closely; I think that it would lead to better maps. Norway shows that sensible application is possible, focusing on getting minimum separations et al right on a hundred objects, and not getting bogged down on, say, three slightly deviating depiction choices seen as truly valuable and legible.
@EricW: I must admit that when I first heard about the Swedish office checking I was very skeptic, but after listening to the WOC2016 presentation I realized that this is just like LiDAR point clouds: It does not guarantee, in any way, that the final map will be perfect, but it does make it easier to avoid some common error sources.
I.e. applied with some consideration it is just one more tool in the pursuit of good, accurate and readable maps, but the map maker is still by far the most important part.
"applied with some consideration"
Agreed Terje, this is the key phrase, but I don't see this anywhere in the IOF language, and all evidence that I can see/hear about MC and SEA behavior has been strongly to the contrary.
Yes, I can believe some. not all, of the the WOC 2016 mapping left room for clutter improvement, but please help dispel the notion that the solutions are technical rather than human. Sure this is a reasonable tool, but certainly not essential, and the real problems lie elsewhere.
The claims and celebration of this quality control process are way out of proportion to its potential and achievement. In fact the SOFT claims strike me rather dubious, mostly an achievement in spin and self back patting, leaving behind some very serious divisiveness, and no real improvement in map quality.
From a broader and historic perspective, the map problems themselves, are quite minor. Maps have never been better, and are certainly not a shortcoming of our sport, yet we insist on more confrontation, with the map makers no less, one of the groups most responsible for advancement of the sport.
@Blair, I ask you again, or any other IOF rep, please give us some IOF evidence that puts this in the proper perspective "applied with some consideration", that Terje/ Norway has described.
I appreciate Council's efforts with MC on the "uncrossable" issue.
I would love nothing better than to have my skepticism shown to be misplaced on this one.
With possible apologies :-) , your silence has not been encouraging.
The maps used for the US champs mentioned by yurets were clearly deficient by modern standards, as there wasn't nearly enough irrelevant ink to slow people down. The result was full-speed orienteering, rather than people standing still with magnifiers trying to read the map. And who wants full-speed orienteering?
JJ, what you are saying is, anecdotally, correct. There was a saying about a certain category of maps, that you need to run full-speed, and then you'd be fine, but don't you dare to slow down and start reading it.
@EricW, what's your recommendation for improving maps and insuring quality, in a way that an international federation can implement?
Also, what great travesty do you see occurring if, say, maps were forced to use the visual/legibility aspects of the spec 100%, without "consideration"? The strength of this angst about actually following the written spec is quite noticable; specific identified problems with this seem few and weak.
I recall there being two North American national events this year with Long on 1:7500 for the elite, 1:5000 for the older, and another with thirty centimeter cliffs. (If an early orienteer accidentally kicks over a feature, is that unfair for later runners?). Coming back to North American orienteering after several years overseas, it was marked how much the sport had turned to magnifier navigation, admittedly a continuation of a long trend. Mappers would probably get more consistent praise if they didn't respond to checks of quality with hysteria. It's just a simple check. I suspect that the IOF checks, and hopefully more national checks, will turn out fine, nay brilliantly.
As far as I know the proposal wasn't discussed with FOC (nor would I expect it to have been since it's largely outside our area of responsibility). It was approved by Council at their last meeting.
Hysteria? EricW was asking that the computer check of the mappers efforts be "applied with some consideration". I totally agree with that and the Norwegian approach that Terje talked about rather than the "Computer says No" SOFT approach.
Well, there seems to be wailing whenever someone suggests applying standards or metrics. I agree that the Norwegian approach seems sensible, but even if the standard were applied exactly, it's not the end of the world, and better than what we have now.
What were the events with 1:7500 for the elite? My first thought was Burnt Mountain, because that did have 1:5000 for some classes, but the elites were on 1:10000. Was Idaho one of them?
Wouldn't it be great if a computer could make just anyone without talent into a great mapper? That would indeed be better than what we have now.
@blair, thank you your public comment. I'm always glad to see it, even when I'm hoping for more insight :-)
@Jim, recommendations? First, we have to lower expectations. There is no way to insure quality. Hell, we can't even define quality. The best we can do is maintain good human relations.
Let's start with the medical principle of "do no harm". This SOFT process, not the OCAD tool, has already been toxic to the mapping community in Sweden, (a very central primary source), and I suspect, but can't verify, with very little change in real map quality, just the test beating willingness of some mappers. Sorry, the current "nothing" is indeed better than the SOFT model. (again, not about the tool)
Mappers are stubborn, including the best ones, and resistant to authoritarian dictates, and especially data outputs. The best chances to teach or modify behavior involve personal, friendly, constructive contact. Clinics, articles, and pleas are OK, but I've always felt project-specific contact with mappers is better, and the earlier in the process the better. Still, don't expect a high success rate here either.
The primary way that organizers, local to WOC, can have an effect on map quality, is through hiring. Do your homework, know your mappers, or consult with those who know more. Investment advice aside, the best predictor here, is indeed past performance.
At the WCup /WOC level this probably means IOF approval of the mappers, probably involving mapping samples, right along with the other sanctions, or simply live with the host country choice (I admittedly don't know the current practice, but guess that it is still a host country decision). I think it is well proven that IOF input after the hiring, however well intended, is more likely to result in heat than light.
Obviously this puts a serious responsibility on the controllers, actually more like the people who recruit/select/ approve the controllers. I don't know who that is, but I can pose the critical question, do we want authoritarian bean counters, or people with better human rapport, who have a little deeper understanding that real map quality is much more than following some tangential specs to the third decimal place.
Identifying and using good mappers is of course very sensible advice. I'm not sure why measuring adherence to the standard is toxic. Nor is map quality indefinable. Every competent orienteer is capable of assessing it. It's central to the sport that orienteers know what to expect the terrain to look like based on the map, and vice versa, including allowing for variations in mapper style and the various ways that something might be depicted. At competition speed. This is a universal, essential, intrinsic, defining capability. And, if hundreds of thousands of orienteers worldwide are to expect the same thing from a map, then it needs to be written down, and titled something like International Specification for Orienteering Maps. Real map quality is matching those expectations, which are rather well captured in that spec. If they're not, and only mappers know how a map should look, then only mappers are capable of orienteering, which rather defeats the purpose. Every competent orienteer with sufficient patience is a potential good mapper. Let's seek good mappers, but not make this too woo. Adhering to feature sizes, symbol sizes and minimum separations is not psychologically brutalising.
"Magnifier navigation" is an outcome of strict adherence to the 1:15k edict.
It shouldn't be. The legibility aspects of the spec should ensure that.
@?? Ken?? No problem with the substance or intent, but having my words edited, (unrelated to decency standards) is a little unnerving, especially considering the "big brother" theme behind this thread.
Another indulgent diversion, but big-brother-techno related, are there any Dilbert fans who see a connection between the recent "health monitor" capers and this thread?
In my professional line of work, we have something analogous to the checker, that looks at our software for things that, while technically allowed by the language, are considered to be poor style, that don't meet our "coding standards". Such violations aren't forbidden, but somebody has to look at them and decide if they are reasonable deviations. That may be a good way to use this tool for evaluating maps. A deviation (such as the ones anecdotally mentioned above) don't have to be treated as prohibited, but each one needs to be individually considered. If there are too many to look at, then you're probably doing something wrong.
The map in the article is "Saltö" outside Strömstad which, according to the club website, was last updated in 2003 and probably originally drawn in the nineties. WTOC was held in the same area on a completely different map and this map sample was presented as a teaser. For some reason the sample still remains as a "stock photo" used for mapping related news on the IOF page.
I am with you Eric - it seems like a reasonable thing to do but is so open to misuse and abuse. I am reminded of a WOC a few years back where one of the event advisors (an IOF MC member at the time) used a similar function in OCAD to arbitrarily remove all cliffs that were shorter than the minimum length WITHOUT any checking in the field. In my experience that sort of stuff happens all too often when you get a controller/EA who doesn't necessary appreciate the finer points of mapping and compensates by hiding behind the strict letter of the law. The rules need to be applied with some flexibility with the aim of producing the best possible map (including issues of accuracy, consistency, generalization appropriate for the level of detail in the terrain, legibility, etc) and that may require some minor bending of the rules which should be discussed and decided upon by the EA in conjunction with the mapper IN THE TERRAIN.
Nixon - are you saying Tampere maps conform to specs but aren't very good or are good because they don't stick to the letter of the law? Any examples?
robplow- I haven't heard that WOC/EA/MC story, but it is completely in line with what I've seen and heard from other WOC and SOFT sources, stories which are all too common.
..and yes, IN THE TERRAIN deserves emphasis.
Mapping is a balance of objectives, but if the mapper knows the specification, then the mapper could make a short list of proposed deviating features, and present that to the controller, with reasons, rather than, say, presenting the controller with a map with a multitude of objects out of spec.
It goes back at least as far as computer-drafted maps. The main mapper for WOC93 decided that the quantity of rock in Harriman warranted the use of three sizes of boulder dots, and used a middle boulder size when doing fieldwork. This was really the first time that such an option was available. It was also the first time that the IOF controllers had the option of rejecting an idea like that late in the game, and the medium boulders were all converted to small boulders before printing.
Last minute changes certainly aren't ideal, given, for instance, that the distance between a medium boulder and another object might have been exaggerated to achieve minimum separation for legibility. Changing it to a small boulder will give a slightly exaggerated appearance of separation of the objects (if the eye can perceive that sufficiently, maybe not). I think that "this can't be properly represented using the standard" should lead to getting the standard modified, getting an upfront OK for a deviation, or not mapping that terrain for that purpose. Yeah, medium boulder would be useful some places I think (like one that I'm mapping now), but I think that compromises are worthwhile for a universally understood map standard (rather than getting used to something odd on the weekend). In my experience, the number of maps for which a mapper thought some special deviation necessary is enormous, and the number that truly warrant this quite few. If we just started with the norm that all maps meet ISOM or ISSOM, and exceptions are for one Middle event every few years, or a mountain marathon, or other particularly unusual event (MicrO?), we'd be way ahead. My impression is that elite orienteering speed has remained fairly constant over the last decades, after earlier decades of improvement, and my suspicion is that mapping of more than is needed has slowed top orienteers as much as better training has sped them up. Orienteer on a legible map from 20-30 years ago to see how much a pleasure it is relative to today's hunched over map reading.
Medium boulders have been permitted since at least 2000. They were not given their own symbol number as such, rather a 'back door' authorization in that it was permitted to enlarge symbol 206 (small boulder) in ISOM 2000. They are still permitted in ISOM 2017 (see definitions of 204 and 205) so long as the 'medium' boulder is no more than 30m from another boulder.
And actually in earlier versions of ISOM (from the days of hand drawing) there was a 20% rule: all symbols were allowed to be within 20% of the specified dimensions. The rule was probably supposed to allow for unavoidable variations due to drawing by hand in ink, but in some circles it was used to justify things like the 'medium boulder': a small boulder is 0.4mm and a large boulder is 0.6mm . A medium boulder of 0.5mm is within 20% of 0.6mm, therefore allowable. At least that was how the logic went.
I've certainly thought of making use of ISOM 2017's medium boulder option in my current map in progress. In most cases, they fall within the proximity limit of other boulders, but in some cases not, and I decided that would be more confusing...intermediate size boulders sometimes showing as middle size dots, sometimes as small dots. In the distant past, I at least considered using the 20% rule in a similar manner.
" my suspicion is that mapping of more than is needed has slowed top orienteers as much as better training has sped them up. "
Here is another interpretation. Spectators and the average travelling orienteer pay for maps. The average age of orienteering public has been steadily rising. One obvious outcome has been a need to print more maps at 1:10k or 1:7.5k. Perhaps another less obvious outcome is that the average orienteer is travelling slower than two decades ago. If the average orienteer wishes the same navigational challenge as they had twenty years ago (perhaps decisions per minute is an informative metric) , then they might appreciate more complex terrain. I may have only mapped about 20 maps, but one trend is clear. No orienteer ever has said to me "you have too much on the map". But I have often been told that [a feature] is not mapped. When I try to explain generalisation and scale tradeoffs, I am met with a blank stare. That might seem an uninformed orienteering public, but this is the public that pays the mapper.. when and if the mapper is paid. So if you are a mapper working on terrain likely to be used for a WC or RC, fine. Stick religously to the standard. But mostly the choice of terrain to be mapped is based upon the preferences of a local and ageing orienteering community. If they want complex gold mining terrain that is also legible, then you would be foolish to stick with the new spec. Map for the market.
"No orienteer ever has said to me "you have too much on the map". But I have often been told that [a feature] is not mapped. When I try to explain generalisation and scale tradeoffs, I am met with a blank stare."
This is one possibility, and one that mappers do sometimes mention. I have however heard other orienteers complain about map details that can only be read back home under a good light, or such. Much of perceptions on this are anecdotal. It would be good to have a poll, or some other way of discerning the overall sense. It may be that there are two markets, one for as much detail as feasible, one for the standard (which can be quite detailed). I wonder if the orienteers preferring a balance toward legibility are quieter, as there's something of an "I can read more detail than though" thing. (Personally, I can read the detail, but find it tedious, and distracting from the point of the sport, in most formats.) Or maybe there are few who want maps more like ISOM (and they're all on the IOF Map Committee). I wish there were more than anecdote and a few forum postings.
Neil, I haven't used or even seen your maps so I can't comment on them. But Jim raises a relevant point here. And I don't think that decreasing the map scale to allow for mapping complex terrain is the same as decreasing the map scale in less complex terrain and then filling in all the blank spots with excess detail just to fill up the map. Or because people say "you left out [a feature]". I have seen numerous maps where there was too much on the map.
Some of the reaction to the "missing feature" complaint might best be handled by setting expectations in advance. I don't know if this is the custom in your area but most O meets in my radius of travel issue numerous pages of "meet notes". One section of this is "mapper notes" where the mapper says something about the terrain and his approach to mapping it. Things like "many rock features, so only boulders > 1 meter are mapped" or whatever the criteria used. And etc for other significant facts.
While your economic analysis might be an "unfortunate reality" I think we need to get folks on board (yes, even those who pay the bills) with the idea of making maps that help our national teams develop the skills needed for success at the WC level. Running fast in terrain is pretty high on that list of skills.
No orienteer ever has said to me "you have too much on the map".
I'm controlling a British Championships next year, and this was the first thing everyone said about the map, including me, the planner, organiser, potential competitors etc. It was a first attempt with Lidar, so no big surprise there.
We got it remapped by Dave Peel, known for the clarity of his mapping*.
Somewhat late to the party, the UK Map Advisory Group noted the same thing and asked to check the map against the spec.
I expect when they see the new map they'll like it as I did. They may have constructive criticism for the mapper, who should take it on board since he's paid to get it right. They may want us to explain why we did some things - and that would be sharing good practice. they may have ideas we didn't think of. Hopefully they wont come back with a load of nitpicky demands. In the unlikely event that they get all officious, and wont justify or discuss it, we'll just work to get them fired - and that would be sharing good practice too.
So I was surprised at how defensive the team were against such a check. I guess Eric is right about people being "resistant to authoritarian dictates". But sometimes the authoritarians are right, and when they're wrong, the more people who know and tell them so, the better.
*Yes, I do know this isn't an unalloyed compliment...
> . I wish there were more than anecdote and a few forum postings.
On this I suspect we can all agree. I have never spent much time on the Scientific Journal of Orienteering and I can only find issues as recent as 2014. The editorial in the 2013 issue seemed rather downbeat about the future of the Journal. It appears to me there are some very interesting questions in the above thread that would make for interesting research. Here are two.
Market segmentations of orienteering participation, aspiration and event enjoyment
Metrics of map complexity and relationship to running speed and terrain complexity
'New Updated Process for Map Evaluation 2018' in Sweden.
Google translate does a pretty good job - karter = mapper
As far as I can see it seems to be taking the 'technical control' away from the central mapping committee and giving it to the controller. And making sure the 'technical control' is started much earlier in the process.
Trusting your reading, this sounds like a step in the right direction.
Step in the right direction depends whether or not you get a controller like the one in robplow's post of December 1.
On the medium boulder topic -- although I'm not opposed to them, I consider the 20% leeway rule to not be a legit loophole. I can see shrinking or enlarging all of the boulders on a map, or maybe adjusting the size of some boulders in a particularly crowded area to improve legibility, but adjusting some up and others down to make two sizes seems like a deliberate misreading of the spec. (I'm not intending to start an argument -- it's not clear that anybody else here feels differently.)
Relax JJ, that rule hasn't been a thing since 2000.
As far as I can see it seems to be taking the 'technical control' away from the central mapping committee and giving it to the controller.
Does this mean there's a "map controller" separate from the normal "event (course) controller". Or are these the same person?
The first two dot points say that a map controller (kartkontrollant) shall be appointed for all Level 1, 2 and 3 events. Level 1 being the highest level the map controller is appointed by SOFT in collaboration with the district mapping officer. For 2 and 3 the district mapping officer makes the appointment.
Presumably that is separate to the event controller.
Please login to add a message.