 |
 |
| Home | Search | About IWAI | Join the IWAI | News and Events | FAQ | Sitemap | | |
|
Home
> IWAI Waterways Discussion Forum
> Topic IWAI-SIG/ Charts Posted by Meelickman
Now that we have a group who are interisted in charting the waterways it is time to have our own thread and leave history behind. IWAI- Special Interist Group / Charts (Will some one ask the pres if it is ok to have a SIG within the IWAI before we are all ran off) Taking an overview, the subject can be broken down into 3 sections 1. Data collection 2. Data processing 3. Data display 3.Data Display To date there has been a lot of chat about this Is it possible to take existing data and make it avalible for dispaly on different plathforms, PC ect ? What is the best way to display existing data, in the future. ? 2. Data Processing What data is avalible ? Where can we make this data avalible to all ? I have have 1. boundry files for Lough Derg and the Mid-Shannon, shore and islands 2. data files for markers, quays and all sorts on Derg all in Irish grid 3 Data Collection Do we need a standard method for this ? Where does data need to be collected ? can data be collected from previous charts as well as in the field ? I have portable surveying kit to collect data in the field Regards Donal
Could I suggest that you have a look at the website created by Captain Trevor K. Northage, MNI, aFRIN at www.anglingcharts.com He includes an excellent overview of chart making in his “GPS Cartography Tutorial” and “How the Charts are Made section.” Captain Northage has produced charts for Lough Corrib, Lough Inagh, Lough Bola, Lough Scannive, Ballynahinch Lake, Chapel Lake, and the Upper Ballynahinch Fishery scheme, and is currently working on Derryclare Lough in Connemara. The Lough Corrib Navigation Trustees are undertaking a five-year project, during which the entire lake will move from the present system to IALA standard marking. It will be done in five stages, starting in the River Corrib at Galway City and heading northwards, to finish at Maam Bridge, Connemara. Lough Corrib is very demanding navigationally. The new marking system will make it far safer and hopefully encourage more visiting boat owners to explore this beautiful waterway. The Navigation Trustees are working very closely with Captain Northage’s charts & GPS, particularly with regard to highlighting the many isolated dangers/shallows that were not indicated on the original British Admiralty chart. Zara Brady Hon. Secretary, IWAI Corrib Branch "Why Not"
See my notes below......!! (Will some one ask the pres if it is ok to have a SIG within the IWAI before we are all ran off) Taking an overview, the subject can be broken down into 3 sections 1. Data collection 2. Data processing 3. Data display 3.Data Display To date there has been a lot of chat about this Is it possible to take existing data and make it avalible for dispaly on different plathforms, PC ect ? What is the best way to display existing data, in the future. ?##### Store the data in ESRI shapefile format. This is the industry standard and allows the data to be manipulated in any GIS program. There is a bit of a learning curve involved with this software, and I'd suggest that the creation of cartographic information from raw data is limited to one or two people. With information stored in this format it will be future proof. Suggest BSB format for present charts. 2. Data Processing What data is avalible ? Where can we make this data avalible to all ?#### The Canadian government make shapefile information available on websites, as do others. It would only be of use to anyone who knows how to use it. However you only want one version to be available of the final charts - not "joe soaps" own interpretation so suggest the shapefile information is not public. I have have 1. boundry files for Lough Derg and the Mid-Shannon, shore and islands 2. data files for markers, quays and all sorts on Derg all in Irish grid - #### need to be converted to WGS84 / and a projection to be agreed, probably a mercator projection for universal compatibility - not a problem. 3 Data Collection Do we need a standard method for this ?### the base data should have a standard collection format if possible. This isn't going to be a professional system, so some flexibility will be required. Assuming its sounding data then a datum will need to be known. The timestamp will be in the file - so all you need is a record from the ESB or IW/Fisheries of levels at different positions. They probably have online data or a database. Where does data need to be collected ? can data be collected from previous charts as well as in the field ?#### yes it can, as long as it can be correctly georeferenced. Old data tends to use different datum, often unknown. There is a section on my website about georeferencing Google chart data. I have portable surveying kit to collect data in the field##### Brilliant. Excellent for producing reference points for georeferencing old data. This would then need to be rechecked physically as part of a survey program, but "will do". Trevor www.anglingcharts.com Mucking about in boats. Edited 2 time(s). Last edit at 23/12/2011 03:21 by Capt Cautious.
Once you have the chart in shapefile format - and have produced, for example BSB charts, there are no end of new apps to run them. [shop.memory-map.co.uk] for example, this app will allow the use of converted BSB charts. I really believe this project is the way forward. Trevor www.anglingcharts.com Mucking about in boats. Edited 1 time(s). Last edit at 23/12/2011 03:32 by Capt Cautious.
Hi Trevor - I'm sure you're aware of the OpenSeaMap sub-project of OpenStreetMap. (Mark Leyden mentioned it on the thread that kicked off this SIG) Wouldn't OpenSeaMap be the natural repository for all data gathered by any community effort? (Hopefully you won't find yourself doing all the work)
Regards, Glen
Hi The problem with open seamap is that it is unauditable. I have a considerable amount of information up on open streetmap already, I did it very early on in the project. I could literally have put ANYTHING on there as long as it looked like a proper topography. Would you trust open seamap knowing that some joker who doesnt have a clue has moved the position of a channel, or marked a safe route where a rock is, maybe marked 40ft instead of 4ft ? The system MUST be auditable to retain creditability, and the contributors need to be part of a coherent team. For producing a pretty map its fine. For producing something which will actually be used in critical situations it fires way off the mark. Its nice to see contibutions on line and I still get a kick out of seeing all my "stuff", even if it could just be a load of old tosh. CC www.anglingcharts.com Mucking about in boats.
Quote:- I see your point, lots of scope for bored teenagers to have fun. I doubt anyone on the inland waterways will be ignoring the markers they can see with their own eyes, but your point stands with regard to depth contours and features that are not physically marked. Quote:- in any official way, or just something like regular official releases from an IWAI mapping team? Regards, Glen
Re auditing chart details. We have a responsible nucleus of interested parties. We would be more than capable of doing our own checking I'm sure, and I'd like to think that the people using the chart would trust the creators to do this properly. All the while the accuracy of chart remains in the interest of the people creating it then we have a solid foundation. As an example - Fred blogs sends in a tracklog with a sequence of unexpected soundings. This would be promulgated to other participants, and when one is in the area he/she confirms the finding. If equipped to do so he conducts a small survey grid, or even just passes it on to somebody who can conduct a proper check. Each new event would have to start a sequence of indexed steps, resulting in an update to the chart. All participants would have access to the list of outstanding items, and as the chart would be open source I think it would be of great interest to many others to be able to see things happening. Not many people get the opportunity to chart their own waters, and its something which invariably only gets done properly just once. Its a terrific way to get to know waters inside out, and gives us all something else to justify mucking about in boats. Trevor www.anglingcharts.com Mucking about in boats.
Unless previous data is made available - then the thing will simply die. I have my doubts that any privately held base chart data will be available from any existing charts. So far nobody has actually sent me a sample of this data so that I can make a "proof of concept" available on the website. If I can do this, then everybody can give it a whizz on whatever platform they care to point it at, before committing to a way ahead. I've already had people asking me "tell me how to do it ?"...but I rather feel the data will remain closed. Fair enough. Someone else did all the legwork, I would not be willing to part with chart base data, even if I was letting the finished product into the wild for nix. But thats just me. Its asking rather a lot of someone, no matter how altruistic they are. It would be a shame to have to start completely from scratch, and a total waste of resources reinventing the wheel. We should probably draw a line under this. Trevor www.anglingcharts.com Mucking about in boats. Edited 1 time(s). Last edit at 23/12/2011 20:12 by Capt Cautious.
iPhone - iPad - Android and PC for all those wanting to try out something that will display the Shannon Charts on their PDA PC or Notebook - I've put a sample chart that runs in Memory-Map available on the Anglingcharts website [www.anglingcharts.com] For want of something bettter to do on Christmas day away from home I've been playing with the Memory- Map program. It is probably an ideal solution for the Shannon Chart, giving the ability to Print paper charts, as well as run the chart on Laptops and PDAs - thus covering the entire community from Luddite to Techno-frog. Memory-Map is available here [www.memory-map.co.uk] I've got no Shannon data to work with - so you'll have to make do with the Corrib. Trevor www.anglingcharts.com Mucking about in boats.
Hi, following up on Donals original post, I see 4 requirements: 1. Data Collection / Gathering 2. Data Storage / Licensing / Availability 3. Data Processing / Map Creation 4. Data Display / Mapping Applications On the 1st: OpenStreetMap (OSM) provides both the base data and the tools to allow collection. Currently, depth and marker locations are missing. Note that all the base data in the OSM database is covered by this license [wiki.openstreetmap.org]. I have done some early work on Lough Ree using their JOSM editor, and added harbour detail etc. using Microsoft Bing Satellite imaging - this has been allowed as an input to OSM. [wiki.openstreetmap.org] NOTE: all data entered into OSM must conform to the OSM license and as such must be from free sources. So, you cannot copy locations of marks, for example, from a copyrighted map. (I was interested to note that there is indeed an existing database of marks with GPS co-ordinates - if we could, freely, get our hands on that!) There is a wiki page that covers Ireland [wiki.openstreetmap.org] and I have created an Inland waterways page also - [wiki.openstreetmap.org] As I see it, OSM covers no.2 above very well. In answer to Trevors point about the lack of auditing in OSM - this is not correct. All data is stored in a database and comes with an audit trail. However, there is a problem in that virtually anyone can change the data. This can (and is) overcome by registering interest in a region, and the system will alert you with all changes and errors in that region. Also, we would need to have a release process for charts which you would expect would pick up major errors. Data is exported from OSM as XML - [wiki.openstreetmap.org] and can be processed into pretty much anything you might want. To get maps on a Garmin - [wiki.openstreetmap.org] To get maps into OpenCPN - [wiki.openstreetmap.org] To get maps onto a Web Server - [wiki.openstreetmap.org] plus many others... As I see it, requirements 1 and 2 are separate from the chart creation and viewing (points 3 & 4). Further, as I see it, we can start using these tools and improve the resulting charts over time. So, for example, we may have no depth information at first - but this would come... Thoughts? Mark.
Hi Mark, the system may well be kept on a database with traceability. But traceability to what or who ? As you said - ANYONE can enter data, this destroys the audit trail as the data must be traceable to a qualified source. There really is no such source, so there is no audit trail. How do you intend to sort the creditable data from the bogus data by simply looking at the data source ? As I said before, the entries for my area on openstreetmap could be entirely fictional, or could simply be misplaced by 50ft. Either way nobody has checked, and the information appears on the map as valid data. Theres not much can go wrong wandering up a bog road that's marked as public access when maybe it's not. The problems likely to be encountered afloat are somewhat different, and less ameniable to trial and error. If there was a way to use and enter this data in a closed community then perhaps you are onto a valid way of accumulating data, but all the while some anonymous lad sitting in the computer lab in the local secondary school can register an interest and contribute total tosh its not going to be trustworthy. Just my thoughts anyway - I use and have to trust charts for a professional living, and would keep open source data strictly at arms length. Trevor www.anglingcharts.com Mucking about in boats. Edited 1 time(s). Last edit at 29/12/2011 13:49 by Capt Cautious.
Apologies for the length of this post... [quote Mark Leyden]As I see it, OSM covers no.2 above very well....we can start using these tools and improve the resulting charts over time. So, for example, we may have no depth information at first - but this would come[/quote] I agree - start with the low-hanging fruit i.e. markers, jetties, harbours. Fairway, hazard and depth info can be added as it becomes available. Could I propose the following approach to start? (obviously it doesn’t have to be the only approach as long as data is freely shared)
[Quote Capt Cautious] the data must be traceable to a qualified source. There really is no such source, so there is no audit trail. .... How do you intend to sort the creditable data from the bogus data by simply looking at the data source?[/quote] Both OSM and PostGIS (a database designed for use with GIS) provide tools to export to most formats, so the storage and auditing of the SIG data is the key concern here. To assuage concerns regarding change control and the ability to audit, the SIG could maintain a website to provide downloads of the current official release of the charts in a printable format, and in exported formats for use in chartplotters and GPS units. The SIG's dataset is maintained separately from OSM, so it retains credibility as long as the data is reliable - it can only be edited by SIG members. If bandwidth allows, the site could also allow users to browse the charts online. In a community spirit (and probably to satisfy licensing requirements) the SIG’s data can be exported back into OSM as part of each release, so the SIG’s data would also be available to browse/export/edit on OSM by OSM users. But the overlay on the SIG website would be editable only by specific SIG users, so it remains the credible, audit-able source. At my present (somewhat limited) level of knowledge of the technology, an approach might be: Milestone 1 (toe in the water):
Milestone 2 (overlay some SIG data):
Milestone 3 (Export to ShapeFile and any other relevant formats):
Trevor’s expertise will be vital here - the projection used for OpenStreetMap tiles is Spherical Mercator (EPSG:3857). This may or may not complicate the process of exporting to ShapeFile? Milestone 4 (introduce weekly refresh of OSM base data):
Milestone 5 (go public):
Milestone 6a (introduce weekly merge of the SIG’s data back into OSM): Milestone 6 could wait until Milestone n is well advanced. 6a and 6b would need to be accomplished at the same time, as we would need to be able to handle community-edited data within a week of introducing the SIG’s dataset back to mainstream OSM.
Milestone 6b (maintain integrity of SIG dataset when refreshing mainstream OSM data):
Milestone n (introduce tiles for the entire waterways system):
Perhaps tiles should be made available only after the minimum acceptable SIG data (e.g. markers and fairway) has been added? It might be useful to announce tiles as they become available to keep the project visible and maintain a sense of progress. If anyone managed to read this entire post, I'd appreciate their thoughts... Regards, Glen
Hi Glen, Great stuff, that seems to be a solution to the problem - I (personally) could do with a slightly more "condensed" version as its rather a lot to take in ...an age thing.....lets call it an idiots guide. Perhaps you could replace "week" with "month" ? I'm sure anyone invloved won't want too much pressure ! Fiddling with charts is a sure way to "disappear" for many , many hours. I'm not sure about the accuracy of Bing projections. Has anybody run any comparisons ? Certainly if the orthorectification is on a par with googles (good detail, but certainly not accurate enough for our purposes) then there is a lot of work to do before the imagery can be used for drawing georeferenced outlines. The errors vary with location and need to be quantified for every area charted. Pop down to the local planning dept - get a load of reference points as ITM Irish transverse mercator or read some positions somewhat less accurately from the OSI site (positions for recognisable points), then convert to whatever projection Bing uses - and compare. On a different note....is there any chance we can set up an email forwarder that we can reply to for this SIG. I will have no html access for the next month ! Trevor www.anglingcharts.com Mucking about in boats.
Quote: Hi Trevor. Agree, QA and safety is a reason one would be careful about involvement in community or opensource marine mapping projects. It's one thing for a car to take a wrong turn because of a map issue, but another matter altogether for a vessel to be put in danger because of a mistake. The broad concept sounds good in theory, but unsure how it could turn out in practice. OceanFroggie
Quote: I note the license agreement for the excellent closed-source project (pre-Garmin-ownership) states that they are for information only and should not be relied on for navigation. And Garmin's disclaimer for their product incorporating the recent closed-source data states "data and bottom information are for navigation reference only, as bottom conditions are continually changing. This product is for general reference. ..... Garmin makes no warranty as to the accuracy or completeness of the map data." And this is entirely reasonable - the charts are for non-commercial, leisure use, and priced accordingly. At present, I'd expect this project's data would be made available under the same disclaimer and license used by OSM. If there are any tame solicitors out there who would like to contribute on the legal aspect, shout. I doubt anyone will be using the charts for night navigation on autopilot! And they wouldn't be intended for that purpose. Nor should anyone blindly venture into shallow/unmarked areas without a depth sounder+due care & attention, or ignore markers they can physically see if they conflict with the proposed chart. It's worth bearing in mind here that this effort is intended for leisure cruisers on the Shannon, not oil tankers in the Straits of Hormuz :-) As for how it'll turn out in practice, I suppose that's the purpose of the first 2 milestones outlined earlier - we'll know at that point whether this approach is feasible or not. If it gives a reasonable degree of accuracy, I think it's worth pursuing. If it turns out to be a lot of work for a poor result, I'll re-think. Glen Edited 1 time(s). Last edit at 30/12/2011 11:52 by GP.
Hi Glen re: It's worth bearing in mind here that this effort is intended for leisure cruisers on the Shannon, not oil tankers in the Straits of Hormuz :-) 1. Precisely. leisure cruisers tend to be FAR more litigous. 2. Its not an oil tanker 3. I also have a leisure craft - and have that particular hat on when I'm writing about charts, it is simply that I am able to draw on a particular skillset, which is transferrable between the two disciplines. I think you'll find that most navigational charts have a small amount of professional expertise involved somewhere in their creation, maybe for a very good reason, maybe not. I'm sure you'll find out. Cheap shots not appreciated. Trevor www.anglingcharts.com Mucking about in boats.
I'm with Trevor and Noel on this one Glen, once a chart becomes available it is assumed correct despite the warnings. And there are times when they would be used at night or poor visibility, mainly to ensure one was in safe water and heading roughly in the right direction, albeit in conjunction with other tools. I had a look at both proposals, and both are equally as good as far as I can see (once the traceability element is covered as per Glen's clarification). But the advantage with Glen's and Mark's approach is at least there is a starting point of outlines, which we don't have for Trevors suggestion. I know the outline data is very rough, but it is a start. Is there any way to remove the County barrier lines, they look like routes, and could be very confusing. If we had data, then I would prefer Trevors solution, but we don't have it, and are unlikely to acquire it. Major markers on the lakes should be easy enough to log, and there aren't that many of them, so a few of us inputting that data should be pretty quick. The Erne would be a different proposition, it is dotted with wee markers all over the place, and that will take some time. Keep talkin lads, Donal will be back soon from his holiers, and we can see what his view is as well, he has some data, but not sure what it is. All views are welcome at this stage, so lets keep it friendly. We are still in feasability but really sucking diesel now. Les 41M
Has anybody actually checked the outline data available in openstreetmap ? Perhaps then someone could trace the contributor and get some sort of provenance ? It may have been walked - canoed, traced from Google or Bing, etc. All will give different results. My personal preference would be for a consistent provider / acquisition method for topographic data. Multiple sources means unpredictable and inconsistent errors. A level of accuracy which is required should be agreed upon, if 10m is neither here nor there then go for Google or Bing, with either of these, if you intend to georeference it properly then outside sources need to be able to accomplish this competently to an agreed projection and datum. There is no point in having data which you have no idea is georeferenced or reprojected - or not. A small team should probably agree methods, and make their own shoreline data using consistent methods. Trevor www.anglingcharts.com Mucking about in boats. Sorry, only registered users may post in this forum. Online UsersGuests within the past 30 minutes:
28 Record Number of Registered Users (within a 30 minute window):
13
on 26 February, 2013 Record Number of Guests (within a 30 minute window):
154
on 08 December, 2012
|
- 
