Quinafloat

Right, The report as I believe it should be, Normalised, human readable and historically accurate to account for change of mooring and change of boat name. Customer number: 123456 Customer Name: Joe Bloggs Boat Index: 37678 Date Invoice No Value Description Boat name Mooring 1/1/2012 12345678 £300 Licence 12M Kingfisher 1/1/2012 12398765 £200 Mooring 3M Kingfisher Little Snoring 1/9/2011 12456789 £200 Mooring 3M Kingfisher Big Snoring 1/6/2011 12456799 £200 Mooring 3M Serendipity Big Snoring *Both the home mooring and the boat name can vary over time. The only stable items are customer number and boat index number. Customer name may vary over time (marriage, deed poll, etc.) but should always be show as the current value as the customer name changing over time is not important to the subject matter. The relationship of the boat to the mooring at any one time is the main basis of paying for a mooring and both should be shown as they were at the time. Obviously a licence fee does not relate to a mooring. I find it hard to believe that the cost of producing a competent report can be cited as a reason for producing an incompetent report. The costs of a court case are massive and any competent coder could produce a such a report in less than a day (provided the data existed, which I doubt) within the database. Naturally such a report could be used for all court cases requiring documentation of financial transactions.

TD is a boater being taken to court and may well loose his home as a result. I don't care if he bites my head off for every post I make. I will continue to offer any help I can. Let this not descend into a personality issue.

Both headings are not headed by Func Loc, on the one report it is Func Loc, on the other it is Functional Loc. I pointed this out in an earlier post. Any expectation that they would contain the same data would be misplaced, otherwise they would have the same naming. Your explanation should read. Func Loc Location of Home Mooring at print date. Functional Loc Location of the Home Mooring on the date and time of uploading to the database. However it is CaRT's task to clarify the meaning of the Meta Data, both you and I could both very easily be wrong.

So, are people saying that TD was sent the wrong report, or that he should ignore the incorrect parts of it without telling him which parts are incorrect?

The reports are an integral part of the system. If they are inaccurate for whatever reason then the system is not fit for purpose. Think of a chain with a broken link, it is not a good chain.

The question of whether the boat logging system is fit for purpose is a composite of many parts. Datalogging integrity, Data upload protocols integrity, database integrity, and report integrity. Overarching those technical points are the periodicity of sightings, granularity of sightings (what physical area is considered the same place) and what level of those 2 items is considered acceptable by the boating public and the courts. Add to that the need to audit the system to ensure that no-one from either side is lying or manipulating data. If all of those points are satisfied then the system should be considered fit for purpose. However it's purpose is nothing more that to provide a factual basis regarding the sightings of boats. It has nothing to do with whether the are there legally, that is for human judgement.

Tony, the names of the data showing home moorings are different in the two reports. This means that they are totally different pieces of data, even though the one was, almost certainly copied from the other at different points in time.

This is a boat with a home mooring, Bona Fide cruising is not an issue.

The boat logging report. Missing data: time of sighting should be included, it is perfectly possible to be logged twice or more when cruising. Long/Lat of sighting Accuracy of GPS at the time of sighting The base location recorded by the GPS unit has been reverse geolocated (translated) into a position (location code) at some point in the processing. The consequence is that CaRT have, de facto, defined place., and as a consequence how far a boat must move to be in a different place. GPS can be quite inaccurate in situations with overhanging trees or buildings close by. Sometimes these inaccuracies can be hundreds of metres. The lack of accuracy data is important. It should be noted as regards home mooring that in the financial report the key of the home mooring has been included as well as the location code, whereas in the boat movements report only the location code has been included. This would indicate, but not prove, that the home location code was added during the boat logging data upload process. Hence it's historic accuracy in this report.

Could someone give me a breakdown of the location codes used on the report? format 1 xx-999-999 format 2 xx-999-999-99 format 3 xxx (shown on report as "MAC") Thanks

Thanks Nigel. I will dig into it.

The report was for financial transactions in relation to home moorings, nothing to do with boat logging. I have never seen a report from their boat logging system. Somehow the two subjects have got a little mixed up in this thread. The report showed, amongst other things that the home mooring did not reflect the home mooring at the time of the financial transaction and gave the appearance that previous home moorings had been overwritten by the current home mooring. Removing the last logged position if the boat is not logged in the same area, nor logged elsewhere, is technically difficult, but far from impossible, which is why I suspect it is not being done. All of the pointers however point in that direction. I do not know whether you could obtain information from CaRT as to the "intermediate" date sightings taken in the same area which do not show your craft. It would depend on their data upload protocols.

In the NBW article it states: Here it should though be noted that once a physical sighting of a boat's position has been logged by a Data Checker, the computer then replicates that sighting until such time as a Data Checker follows up with a further sighting. Whether that covers your point(s) or not I am not sure. It would indicate to me that not having a further sighting would continue to replicate the original sighting, When, some months later, a further sighting takes place in the same area, without intervening sightings the system considers that your boat has not moved. There was a case recently reported on London Boaters facebook page of someone receiving a notice after returning to an area after some months of absence in just these circumstances. Luckily the boater could prove where they had been and when. (London Boaters, July 14th). Unfortunately most of the sparse knowledge of CaRT's boat logging system is gained from how it reacts rather than it's specification, I have asked for a meeting to with CaRT to align CruisinLog's GPS readings with CaRT's dataloggers to enable easier cross checking in the event of disputes but I never received a reply. It should be noted that a GPS position is not an absolute position, a GPS unit gives a position and it's level of accuracy, which is why I refer to an area.

It does seem to happen that boats are considered to be in the same place unless logged somewhere else. There was a recent report of such an incident on the London Boaters forum. The link provided by Alan de Enfield http://www.narrowboa...inuous-cruisers above would seem to verify that the situation over boat position logging is that a boat is considered to be in the same place unless logged elsewhere. There are many ways that the boat logging system could/should be improved but if the system described in the NBW link is the one in use, then people need to be very aware of it.

That is why I wrote CruisinLog. I could get sighted at Gayton Junction.I then overwinter on EA waters for 5-6 months and then if I get sighted at Gayton on the way back I would be on CaRT's hit list as not having moved for 5 months. CaRT don't have dataloggers on EA waters despite issuing Gold licences. This is not theoretical, This is the trip I make 2 years out of 3.

No worries, the IT problem being discussed is the systems analysts worst nightmare, how to ensure correct historical context is being maintained. The problem is not where TD's mooring is, but where it was some years ago when it was in a different place.

It would seem to me that a financial transaction for a mooring would (among many other things) be intersection data between a boat record and a mooring place record. At the point of creation of the intersection data a copy of both the (then) current boat name, and the (then) current mooring place key should be stored on the intersection data record as either entity record could change over time. This would allow a historic query, such as the above, to show valid data in that respect. Whether this is sloppy systems design, or sloppy query design is the point at issue. If the consideration of historical queries has not been taken into account then no amount of data normalisation would have helped as the attributes of the intersection data would be short of those fields. It is, after all, part of an analyst's responsibilities to ensure that what a system reports is both correct and intelligible. That is not the situation here. The current location is being shown in association with a date where it is not valid. I agree with you that, if the current location is being shown, and if it is the only location being shown, it should be shown separately clearly labelled as what it is. I also agree with you that such a report could be produced with valid underlying data. The thing is that if the underlying data is valid, then producing such a report is misrepresenting the facts contained in the data.

Tony, your boat movement records may well have been recorded by one of their datalogger machines and will have been recorded as Longitude/Latitude rather than location name. Possibly even a mixture of the two. If you need any help interpreting the Long/Lat data to an exact physical location just PM me with the data and I will plot it out for you. If they have place names and no Long/Lat coordinates then ask for them as reverse geolocation (determining place names from Long/Lat coordinates) is a black art with no guarantee of accuracy in the result unless in an urban environment. You would need three items for each position, the Longitude, the Latitude, and the accuracy of the GPS fix.

Tony, this report is in all probability an Ad Hoc report for management use, and management may well be aware of it's shortcomings. It is the base data, archived original documents (if they exists) that should be obtained that should show the correct location. If those archived documents do not exist then your case is strengthened that their systems are not fit for purpose. It is possible also that the person who created the report selected the wrong location data. That is the current location rather than the financial transaction location (if such data exists), hence the need to clarify the exact meaning of the headings if this document is important to your case.

Provided that your comments on the mooring years are correct, then the report is quite obviously telling porkies. My guess is that the relationship between the boat and the mooring has been changed at some point and the true historic situation has been lost. The earlier post by TD about his experts findings look, at face value, to be perfectly correct. I do not understand the headings. FunctLocDescript and Func. Loc. (if I am reading that right, it's a bit fuzzy) could mean anything that the author of the report wishes it to mean. A full English description of the names assigned to the data should be requested. My best guess is simply sloppy systems design. The alternative (which would support MtB comments) could be that the query did not check up on available history. It would be good practice in any financial reporting system that an exact copy of any financial transaction is retained. This report looks to be a composite of financial transactions and what the author of the query presumed were static data such as location and boat name (neither has any reason not to change in reality). It is quite possible that independent, paper /fiche/image records have been archived which show the true transactions. These should also be requested.

In order to win or lose your money you would need to know the same answers as TD is trying to find out. What is/was the specification of the processing of that transaction type and does the actual processing actually follow that specification. I very much doubt that historic links between home mooring entities and boater entities are retained as this would would severely complicate any query process unless the most current link was, in some way, used as a default. Personally, if I were designing such a system I would record such changes to the transaction audit trail and hold a single current copy of the relationship. Do you know that a relational database is in fact being used, not that it is particularly important? Virtually any database or file organisation that is supported by SQL or similar query facilities would/should react in much the same manner. Of course the main question is, how much money are you wanting to lay on your prognoses? I might be interested.

Yes, a computer program cannot normally correct crap. However the crap frequently comes looking neat, tidy, and with suitable headings that make it look as if it's correct and is actually information. Also computers tend to push out a lot of data at one time making it very tedious to go through line by line. All computer output should be treated as suspect, regardless of it's source.

Paint edges with epoxy resin.

I doubt that it was the intent of the 1995 act that people should return to their home mooring after cruising otherwise the act would say so, but it was probably the expectation that people would do so, and in reality that is what happens. This leaves the question of when a cruise starts or ends. That would seem to rest entirely with the boater as it is not defined in the Act. It is only when the boater deems that a cruise has ended that it actually ends. Similarly a cruise starts when the boater deems it so. If a boater with a home mooring goes on a cruise for e.g. 3 years I do not see that the boater is doing anything wrong. If they return to their mooring and start a new cruise the next day they are surely, perfectly entitled to do so. The bit about about boaters returning to their home mooring to to ensure that there are mooring spots for people cruising is just a pure red herring as that was also not part of the Act. It is no more than coincidental that it is a consequence of boaters returning to their home mooring, as most do regularly.