Re: Database Anomaly

John Pumford-Green
 

Hello all,

Sorry for not replying sooner, I don't check the messages very often these days!

It looks as though a typo crept in to YaDDNEt when I was adding Vestfold to the list of "specials" and I accidentally used the same callsign, LJ2B, for both Vestfold and Karmsund.

I've corrected the coast station name for Vestfold:

Vestfold College LJ2P

this means all new reports of MMSI : 258999039 will be given the correct name/callsign.

and also done a correction on all existing records in the main database so that all old reports are also now correct.


The mechanism for giving names to Coast Stations in YaDDNEt is:

YaDD sends each DSC message across the internet in a simple format:

[IO75-02];2187.5;SEL;477192100;SAF;002320004;TEST;NOINF;--;--;ACK;ECC 115 OK

Only the MMSI is sent from YaDD to YaDDNEt.

This allows each individual user to edit their own local Coast Name file in their copy of YaDD, to suit themselves, without impacting on how the Coast Stations are displayed in YaDDNet.

Each incoming message is inspected to pick out the TO and FROM MMSIs.

Normally a Coast Station is detected by the first two MMSI characters being zero... and Ships are all other stations, where the first two characters are NOT zero.

Unfortunately there are now many land-based stations that use non-coast station MMSI, and we have agreed to treat these as Coast Stations.

I now have to test:

is it a "Special" coast station (one that uses a Ship MMSI but is listed as a Coast Station)?
is it a "Normal" coast station (one that uses 00xxxxxxx)
is is a "Normal" ship station (all other cases)

I keep one database table with all the "special" coast stations:

+----+-----------+-----------------------------+
| ID | MMSI | COASTNAME |
+----+-----------+-----------------------------+
| 1 | 219015591 | Sailor Examiner XPE3307 |
| 2 | 237673000 | Piraeus JRCC |
| 3 | 237673100 | Piraeus JRCC |
| 4 | 219055000 | Skagen Skipperskole XPA6880 |
| 5 | 228040600 | ANFR Test Donges FIIR |
| 6 | 258999039 | Vestfold College LJ2P |
| 7 | 235899980 | RNLI College Poole |
| 8 | 219000333 | Skaw Star XPB3047 |
| 9 | 258999189 | Serco Test LJ8C |
| 10 | 219000559 | Polaris Aalborg XPC5421 |
| 11 | 219000022 | Procare Test XP8100 |
| 12 | 219016306 | Simac-1 XPE3817 |
| 13 | 219020185 | Simrad Test XPE6606 |
| 14 | 235899982 | RNLI College Poole 2 |
| 15 | 258999089 | Karmsund Vid Skole LJ2B |
| 16 | 258999139 | Kristiansund |
| 17 | 258999209 | TMS-C Tromso LJ3J |
| 18 | 258999229 | Uit/LJ4 |
| 21 | 264999908 | CMU School |
| 22 | 412000412 | Tianjin Seaman School |
| 23 | 219000225 | Slipshavn MHV Skole XP6351 |
+----+-----------+-----------------------------+
21 rows in set (0.01 sec)


This is just used as a quick way of checking a SHIP MMSI against the known list of "special" coast stations.

If the MMSI is found in the "specials" table YaDDNet knows not to treat is as a SHIP, but to treat it as a COAST. It then looks up the Coast Name from the main Coast Station database table, which has all coast stations (normal AND special).


You can see the current Coast Station lookup table represented as a web page:

http://gm4slv.plus.com:8000/pages/php/test/mmsi_coast.php


I'd made a mistake when creating the "special" entries for Vestfold, and somehow gave it the same callsign as Karmsund.

Hopefully this has resolved the problem.

I haven't been keeping up to date with recent activity, and might have missed other discoveries of land-based ship-station-MMSI "specials". If any more have come to light, beyond those in the list above, can you let me know, and I'll add them to YaDDNet.


On the wider subject, of naming coast stations....


Naming of Coast Stations is not a simple matter, especially nowadays, as more and more systems are amalgamated, and networked together. It's impossible in many cases to make a simple correlation between an MMSI and a "Coast Station". Many people have their own valid reasons for assigning specific names to specific MMSIs, but there's not one correct answer, so YaDDNet's names are MY preferred names.

All we really have is the MMSI, and that's really all we can claim to know about the signal. Anything else (name, location etc) is open to conjecture in many cases.

We "collect" MMSIs.

We like to give these MMSIs a more "human" name, to give us a sense of the source of the signal, but the MMSI is the only thing we really have.


Cheers,

John

On 12/01/2019 23:46, david via Groups.Io wrote:
Hi Rich,
You made my point for me. It is the YaDDNet database with the anomaly!
And that matters less to me than finding out what the 'LJ2B' is for, in two notionally separate locations.
BR, David
-----Original Message-----
If I could shove in my two-penny-worth
As a station's name doesn't form part of the DSC message, in theory
there's nothing to stop you calling it anything you want (within
reason, I suppose!) if you keep the info to yourself.
Once you start sharing, that's when problems could occur Whose
database do we use as 'the one we must all adhere to' and why? One
man's 'Sailor Examiner' could well be another man's 'Pandrup Test',
for example.Same MMSI number, different name. My own coast station
database for DSC Decoder shows the UK coastguard stations with their
names plus a simple "CG" suffix, and shore stations using ships' MMSI
numbers have an asterisk at the beginning and end of their names so I
can spot them easily when I'm looking at the DSC decoder screen. A
problem would occur if that was how YADDNet read it as someone else
might well have the same station with a different name, e.g 002320011
is Solent CG to me but it could well Solent COASTGUARD to someone else
and could result in duplicate entries which would take up oodles of
unnecessary space on the server. This duplication doesn't happen as
John has got some trickery installed in the heart of YADDNet to ensure
that irrespective of what your database may say and what gets uploaded
to YADDNet, it will be crunched saved and output to your own RWW
report using the station name from the YADD database, not yours. The
databases my well be the same but this is done just in case they're
not. Clever dot com
I've no idea what happens when my weekly log gets uploaded to REU. I
just 'send and forget' !
As for the LJ2B callsign....PASS !
Cheers
Rich

Join dsc-list@groups.io to automatically receive all group messages.