LOGIN

RSS Facebook Twitter YouTube
GLOSSARY       

SEARCHGLOSSARY

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

PROFILESEARCH

Avatar

Please consider registering
guest

sp_LogInOut Log In sp_Registration Register

Register | Lost password?
Advanced Search

— Forum Scope —




— Match —





— Forum Options —





 

Minimum search word length is 4 characters - maximum search word length is 84 characters

sp_Feed Topic RSS sp_TopicIcon
Species coding systems in the Knowledge Base
January 6, 2013
12:33 pm
Avatar
Matt
Málaga, Spain
Admin
Forum Posts: 8239
Member Since:
June 13, 2011
sp_UserOfflineSmall Offline

I'm intending to add some new Schistura profiles today, and among them we have photos of a number of unidentified species.

How would everyone prefer these, and indeed any other species in need of identification, to be listed going forwards?

I know in the past we've used thing like 'ST' numbers for Stiphodon and 'PAN' for Pangio, but am anxious not to create any unnecessary confusion since the site is a bit more well known now.

There seem to be 3 three basic options:

- use a coding system. This could be genus specific as we've done before, or maybe genus followed by 'SF01', 'SF02', etc.

- just use numbers, i.e., Schistura sp. 1, Schistura sp. 2, Schistura sp. 3etc.

- use trade names, cfs and affs wherever possible, so S. sp. 1 would become S. sp 'ring', for example. Some kind of numbering system would have to be used for those without any such designation, however.

So, any preferences or opinions?

Cake or death?
January 6, 2013
3:24 pm
Avatar
mikev
NYC
Community Helper
Forum Posts: 1134
Member Since:
January 11, 2011
sp_UserOfflineSmall Offline

Trade names are inherently unreliable and highly mutable... imho they only should appear in footnotes... I think code/number them.

How certain are you of being able to classify an unknown fish into a particular genus? Can Schistura sp. 5 turn out to be a new aconthocobitis later?

If there is such a danger perhaps one should consider the 4th option: number or code all unknown nemacheiline loaches together, with a footnote in each entry indicating the current thinking about the appropriate genera.

January 7, 2013
9:17 am
Avatar
Matt
Málaga, Spain
Admin
Forum Posts: 8239
Member Since:
June 13, 2011
sp_UserOfflineSmall Offline

Ok, so we should just number them 1, 2, 3, 4, 5, etc. rather than using any letters?

Wouldn't have thought the chances of an unidentified fish being moved to a different genus later would be any greater than for those already assigned to a given genus - just look at what happened to Puntius recently for a good example.

Also, if the fish had no genus name assigned to them, how would users find them in the Knowledge Base?

Cake or death?
January 7, 2013
9:39 am
Avatar
Colin
Veteran
Forum Posts: 791
Member Since:
January 27, 2012
sp_UserOfflineSmall Offline

If you used the system with the SF numbers, does SF stand for "Seriously Fish"? I think that's a pretty good idea because then if the fish is referred to as Puntius SF03 anywhere else then the origin of the ID is in the name. A bit like Corydoras World for the CW numbers and you know where to go if you need more info on that fish... i.e.  you ask Ian about CW numbered fish.

 

I guess each genus would need to start at SF 1 though as opposed to the CW labelling that covers Aspidoras, Brochis and corys?

 

 

January 7, 2013
11:12 am
Avatar
Jakub
Lutterworth, UK
Member
Forum Posts: 41
Member Since:
December 29, 2011
sp_UserOfflineSmall Offline

Genus followed by 'SF01' sounds good to me. The letters would be an instant reminder of how the designation came into being. Plus, as Colin said, in some cases it may spread. SF gains credit ;-)

Trade names and such could be listed somewhere at the top of the article, or indeed footnote, as suggested above.

January 7, 2013
12:50 pm
Avatar
Matt
Málaga, Spain
Admin
Forum Posts: 8239
Member Since:
June 13, 2011
sp_UserOfflineSmall Offline

Thanks for the great feedback everyone and yep Colin, that's basically what I was thinking, with each genus starting at number 1.

If we go with that system should I go back and relabel things that have been coded differently in the past so that we're consistent across the board?

Cake or death?
January 7, 2013
3:29 pm
Avatar
Plaamoo
Bellingham, Washington U.S.A.
Community Helper
Forum Posts: 1253
Member Since:
March 15, 2009
sp_UserOfflineSmall Offline

If we go with that system should I go back and relabel things that have been coded differently in the past so that we're consistent across the board?

 

However you choose to do it I'd say consistency a good idea.

January 8, 2013
10:36 am
Avatar
Matt
Málaga, Spain
Admin
Forum Posts: 8239
Member Since:
June 13, 2011
sp_UserOfflineSmall Offline

Having slept on this not sure how comfortable I am with the idea of 'SF' numbers. Don't see what value they add to anything beyond referring people back to ourselves which seems a somewhat selfish approach, unless I'm missing something?

Perhaps a simple numbering system would be best?

Cake or death?
January 8, 2013
3:26 pm
Avatar
Plaamoo
Bellingham, Washington U.S.A.
Community Helper
Forum Posts: 1253
Member Since:
March 15, 2009
sp_UserOfflineSmall Offline

Matt said
Having slept on this not sure how comfortable I am with the idea of 'SF' numbers. Don't see what value they add to anything beyond referring people back to ourselves which seems a somewhat selfish approach, unless I'm missing something?

Perhaps a simple numbering system would be best?

Your humility is admirable Matt, but if the system originates from SF, shouldn't it be clear that's the case? It seems only practical to me.

January 9, 2013
9:20 am
Avatar
retro_gk
Member
Forum Posts: 100
Member Since:
May 28, 2009
sp_UserOfflineSmall Offline

sp. 001... is the simplest and most consistent form. If a collection locality is known, or if the fish has other names (trade or hobbyist), sp. "location" and alternative names should be mentioned somewhere and, if possible, included in the meta/search tags for the page. This is for the benefit of search engines.

That said, an SFnum label is a good idea. Publicity is always a good thing with online media :D

January 9, 2013
11:03 am
Avatar
Matt
Málaga, Spain
Admin
Forum Posts: 8239
Member Since:
June 13, 2011
sp_UserOfflineSmall Offline

Ok cheers lads. Smile So we go with SF numbers then, in the format genus-SFnum starting from 001 in each case?

Rahul, we've just started the process of adding that kind of info to the search tags for each species, including synonyms, but it'll take a wee while to get through them all, not least because there's a certain loach paper to deal with at the minute. Yell

Cake or death?
January 9, 2013
8:44 pm
Avatar
mikev
NYC
Community Helper
Forum Posts: 1134
Member Since:
January 11, 2011
sp_UserOfflineSmall Offline

Wouldn't have thought the chances of an unidentified fish being moved to a different genus later would be any greater than for those already assigned to a given genus – just look at what happened to Puntius recently for a good example.

I'm more concerned about the possibility when you cannot make a good genus guess based on photos, this will surely happen.

Also, if the fish had no genus name assigned to them, how would users find them in the Knowledge Base?

Off the family. There should be a page that lists all the genera for say nemacheiline loaches, the same page can also list all the "unsorted" entries. The problem of how to find in the KB hits both ways: if you don't know the genus, you cannot find the entries, but if the genus was entered incorrectly in the KB, you will not find it either.

January 10, 2013
8:27 am
Avatar
Matt
Málaga, Spain
Admin
Forum Posts: 8239
Member Since:
June 13, 2011
sp_UserOfflineSmall Offline

Mike, remember a few months ago we talked about adding galleries for unidentified species?

Well, we had a look at it and it really wouldn't be easy to implement with the way the site is designed, plus the fact Wordpress is still quite limited in some ways.

I'd love to include indexes of genera, etc., but think this (and a lot of other ideas) might have to wait for the next major update, so for now we'll have to work with what we have.

You can search by family using advanced search on the main KB landing page, so guess any fish we can't agree a genus for could still be located that way. Oddly though, that's yet to happen in 5 years...

Cake or death?
January 10, 2013
4:45 pm
Avatar
mikev
NYC
Community Helper
Forum Posts: 1134
Member Since:
January 11, 2011
sp_UserOfflineSmall Offline

Yeah, having indices by genera (and in fact as many views on the hierarchy as possible) would be nice.

Two more thoughts on labeling undescribed species: (1) notice that c* and l* numbers are not based on genera, in some cases the genera is iffy anyway (this is more for l* numbers). However, (2) i'm no longer sure that what I suggested is workable with the way Kottelat reshuffled the loaches. We now seem to have beaufortia and sinogastromyzon in different families... (groan)

Whatever you can do....

January 10, 2013
7:14 pm
Avatar
Matt
Málaga, Spain
Admin
Forum Posts: 8239
Member Since:
June 13, 2011
sp_UserOfflineSmall Offline

Yep. :) That's one reason why using genus-plus-number seems to make sense.

Another is that it avoids the use of lots of alphabetic characters which could become confusing in no time, and it would be consistent across the site. Exceptions would have to be made regarding the catfish and Apistogramma systems already in use, though.

What's best to do with species we're unable to confidently place into a genus is trickier, agreed. More input please!

Cake or death?
January 11, 2013
5:27 am
Avatar
mikev
NYC
Community Helper
Forum Posts: 1134
Member Since:
January 11, 2011
sp_UserOfflineSmall Offline

Upon more thinking: from the point of long-term maintenance genus-plus-number is a bad idea, it will create lots of corrections down the road. Genera change all the time, and we know that more changes are coming, and you will end up having to correct the undescribed species on top of having to correct valid ones. The most efficient system would be *very rough classification*+number, for instance loach+number. If you have a good guess on which genera this may belong to, say it in the comments, but don't include the guess into the database structure. So say H01 becomes SFloach001, Sew01 is SFloach002, Sew02 is SFloach003... SFloach001 would remain a valid entry even if Homaloptera is broken into several more genera as will likely happen. This way the master lists of undescribed species are very few and the only maintenance will be changing the comments with guesses on what these species may be. It will be perfectly fine if there are no guesses on some entries too. And it would be possible to have an index file listing all undescribed entries in each group.
I think this will save you time and actually make a more useful database.

January 11, 2013
8:18 am
Avatar
Rüdiger
Brunswick / Germany
Veteran
Forum Posts: 349
Member Since:
May 2, 2012
sp_UserOfflineSmall Offline

Hi guys,

I have kept my trap shut so far because taxonomy (like many other real scientific disciplines) isn't exactly my thing. My reason for speaking up now is Mike's last suggestion. I understand it and it makes perfect sense to me, which I believe is a good indicator that any regular user should be able to use that system when searching for an undescribed species. Just a thought.Smile

Regards

R.

If you must insist on living in a box ...... at least do your thinking outside.
January 11, 2013
9:39 am
Avatar
Matt
Málaga, Spain
Admin
Forum Posts: 8239
Member Since:
June 13, 2011
sp_UserOfflineSmall Offline

So to clarify, following Mike's latest suggestion, even species that we know the genus for, e.g., undescribed Sewellia, Stiphodon, etc, become 'SFloach-number', SFgoby-number', etc.?

Cake or death?
January 11, 2013
5:24 pm
Avatar
mikev
NYC
Community Helper
Forum Posts: 1134
Member Since:
January 11, 2011
sp_UserOfflineSmall Offline

The thing is that there are few if any species we know the genus for! Sewellia comes as close to an understandable loach genus as possible but it remains quite possible that once DNA data is obtained for all known species it will also get reshuffled, either by a split, or by merging to closely related beaufortia, or both... not saying this will happen, only that we don't know, and it will take decades to find out. Homaloptera and Schistura otoh are clearly still 'garbage' genera and we can be certain they will change and many times. I never looked at the situation with Stiphodon's (was there any phylogeny work done?) but I'm skeptical about being able to say "we know" about anything.

Look at the rainbows example, Unmack et al's paper. It shows much deeper level of understanding than we see with loaches and most other type of fish -- in part of course because rainbows have little variability and were already studied extensively. There are many official species, plus many unofficial with names like Melanotaenia sp. xxxxx ... well, all these names are wrong because the paper shows clearly that the current division of Rainbows only Melanotaenia, Glossolepis and Chilatherina is erroneous. So when Matt gets to rainbow profiles, don't use "Melanotaenia sp. xxxxx" even if such names already reached the hobby, SFrainbow+NNN is the only safe way.

January 14, 2013
4:58 pm
Avatar
Matt
Málaga, Spain
Admin
Forum Posts: 8239
Member Since:
June 13, 2011
sp_UserOfflineSmall Offline

Ok, but then how do we choose names for these codes? Using families?

Think I prefer SF-characid-number than SF-tetra-number for example.

Some names will also appear clumsy under this system: SF-livebearer-1 doesn't exactly trip off the tongue, for example, but neither does SF-poeciliid-1.

Don't think we should necessarily hold back on attempted identification of a fish due to changes that may or may not happen in the future, or because a genus is a poorly-understood catch-all, either.

We cannot say 'we know' 100% of course, but we can make informed guesses and decisions based on current taxonomy. The unidentified Pangio in the database are all clearly members of that grouping as it's recognised today, for example.

Under the new system they may appear as SF-loach-2, SF-loach-27 and SF-loach-396, and I struggle to see how that's a simple system to use. How will users find them?

Cake or death?
Forum Timezone: Europe/Paris

Most Users Ever Online: 246

Currently Online:
1 Guest(s)

Currently Browsing this Page:
1 Guest(s)

Devices in use: Desktop (1)

Top Posters:

Stefan: 1567

Plaamoo: 1253

mikev: 1134

Malti: 1099

Mark Duffill: 1012

Member Stats:

Guest Posters: 0

Members: 30288

Moderators: 0

Admins: 2

Forum Stats:

Groups: 4

Forums: 10

Topics: 4595

Posts: 36615

Newest Members: Brawny, qchris87, grayergroove, bhisma wildan, costeacristianalex

Administrators: dunc: 1323, Matt: 8239