Better than L-Nos. (Shane made me do it)

[Jools]

All,

So, we all know l-numbers. But all they give us are a moniker, a pidgeon hole and an excellent marketing opportunity. Is there perhaps a better way "pre-linnean" way of classifying them ahead of scientific description?

What do we need to know about a fish? Where is it from? How is it kept? What to feed it? How about Panaque sp. LO110O? Where L is Loricariidae, O is Orinoco drainage, 110 is it's maximum size and o means it is omnivorous. Maybe there are other things around export frequency or the issue of DATZ it appeared in. This is not a technical discussion, but one to investigate the utility of such a system?

Thoughts?

Jools

[Cristoffer Forssander, posts: 564]

Very interesting. I would perhaps prefer to refer them to the rivers they are from. Leporacanthicus cf galaxias orinoco e

And to add a jy

[Suckermouth, posts: 1514]

For it to be useful as a naming system, names need to be relatively short, relatively memorable, and relatively stable. For stability, I would argue against including length, diet, or anything else that's going to change based off of future evidence. The L-number system is eloquent in its simplicity. While I think that biological information is useful, you should keep it out of the name. "Oviraptor" was named because it was thought that the dinosaur was stealing eggs, when in reality it was more likely brooding eggs! Scientific names don't get to be changed, but what if we find out a fish gets bigger than we thought? Should we change the name? What about if we don't actually know it's diet? Despite Mats's list on loricariids, most of those diets have not been actually observed in wild fish.

My two cents.

[mummymonkey, posts: 384]

You're a database guy Jools, you know data should be normalised. The L number is the unique identifier and must never change. All other data, origin, size etc. can and do change so must be kept separate. Some kind of on line catalogue would do the trick.

[Acanthicus, posts: 439]

Hi,

another, new system wouldn´t work I guess. Even the LDA-numbers didn´t work as good as the L-number-system. If you start a new system by renaming all the known L-numbers nobody will get into it. I think the golden time of new l-numbers is over and no other system is needed.

[MatsP]

And whilst Planet Catfish is undeniably the main source for a lot of information elsewhere on the web, would it really have enough influence to make the system accepted elsewhere. I don't see Aqualog or DATZ "buying it".

And whatever system is used, I don't really see how it will improve on the L-number system in a way that makes all the work effort and explanation that we'll have to produce time and time again to people who are used to the L-number system (and surely the C, CW, LDA and whatever other systems are widely used elsewhere).

--
Mats

[Shane]

Guilty as charged. I have been instigating to do away with L Numbers as an identification system for years. They just don't make sense any more and often confuse things horribly.

Just today there were two postings in "What is my catfish" asking about L27. Which L 27? Panaque nigrolineatus? P. cf nigrolineatus? Panaque armbrumsteri? P. cf armbrusteri "Xingu," P. cf armbrusteri "Tocantins"? Wait, maybe they mean L 27 A, L 27 B, or even L 27 C? But wait L 27 C is also L 330 and LDA 63, and L 27 is LDA 77. Really, this is a good system? And L 27 is one of the less complicated ones.

I get that L Numbers have become established common names. I am hardly advocating deleting all L Numbers from the site. That would be massively counter productive. However, we do not classify any other group of fishes on PC by common name, so why are we doing it with L Numbers when it causes nothing but confusion?

I could go on and on about this topic and the myriad of reasons to abandon L Numbers as an identification system, but will let others weigh in.

-Shane

PS I am also aware that PC is partly to blame. The acceptance of L Numbers would have died out quickly in the English speaking portion of the hobby if it were not for PC. There is a good reason no new L Number book has been printed in English in over 10 years (there have been translations of German works), and that is because PC has had that market cornered.

[racoll, posts: 4473]

Sorry to be negative, but I have to admit that I don't like the idea. Technical problems with URL persistence etc aside, it's just too complicated for an "average" user. What comes to mind is "created for people who don't know how to use it, by people who don't need it".

L numbers are so ingrained, it will be impossible to convert people over. Just look at how L numbers are now just accepted common names. Peckoltia compta or the leopard frog pleco for example; very few people use these names, they just call it L134.

[racoll, posts: 4473]

Is there perhaps a better way "pre-linnean" way of classifying them ahead of scientific description?

Well there is an official way of doing it scientifically, using the BIN system (DNA Barcode Index Number). Species are genetically clustered, and given a unique species-specific ID (i.e. a BIN), thereby creating an interim taxonomic system with resolvable identifiers.

Here's an unidentified and possibly undescribed Rasbora I found in the trade: BOLD:AAU5133. If this species crops up in collections from the wild, the specimens can be quickly matched to the one I collected, and a name can be applied when the species is described.

[MatsP]

By the way, Jools, did you purposely pick something that contains only L, O, 1 and 0? The guidelines when programming (naming variables) is to avoid using 1 and lowercase l in places that can be confused, likewise uppercase O and 0, because they can easily be misread as the other. [Sorry, that may be a "technical" comment.

I've long agreed that the L-number system is not perfect, but it's a whole lot better than many other alternatives.

And to emphasize racoll's point - in a recent posting on Facebook (you know, the scientific pinnacle! - reply to "pleco suitable to keep with Frontosa - not bristlenose" [not an exact quite of the question]):

Iv got a few in with me tropheus mate L200 ..L27 3 L155 L91

(I did change the lowercase L's to upper case ones - otherwise quote is unchanged, as it makes it easier to read).

--
Mats

[matthewfaulkner, posts: 289]

Just today there were two postings in "What is my catfish" asking about L27. Which L 27? Panaque nigrolineatus? P. cf nigrolineatus? Panaque armbrumsteri? P. cf armbrusteri "Xingu," P. cf armbrusteri "Tocantins"? Wait, maybe they mean L 27 A, L 27 B, or even L 27 C? But wait L 27 C is also L 330 and LDA 63, and L 27 is LDA 77. Really, this is a good system? And L 27 is one of the less complicated ones.

Dude, you are so not up to date This is how it goes, I think.

L027 = Panaque cf. armbrusteri 'Tocantins'
L027a = L330 = Panaque cf. nigrolineatus = Watermelon
L027b = L027c = LDA063 = Panaque cf. armbrusteri 'Xingu' = Platinum
L027c = Panaque cf. armbrusteri (I believe it's cf. nigrolineatus) 'Araguaia' = Goldline
L027c = LDA077 = Panaque armbrusteri = Thunder

And that's just the Brazilian ones (except L330), then theres;

L190 = L330 = Panaque nigrolineatus = the Royal pleco
L191 = Panaque cf suttonorum/nigrolineatus/titan = Dull Eyed/Broken Line
L418 = Panaque titan = Peru Green/Shamper/Shampupa

And then you could go into the Panaque schaeferi/bathyphilus;

L090 = Panaque bathyphilus = Papa Panaque = Small eye
L090a = Panaque bathyphilus = Ojo Chico
L090c = L203 = Panaque schaeferi = Titanic = Big eye
L090e = Panaque bathyphilus = Juanjui

[racoll, posts: 4473]

I would say that making things more complicated than they need to be is not always a good idea. L numbers, despite their flaws, are elegantly simple, and are consistent with described species: they have a unique identifier like the species binomial (the L number code), a "holotype" (the original photo), and a "type locality" (the original published river). Therefore, with the help of databases such as the cat-elog, they can be treated just like described species and easily synonymised when new information comes to light.

The oft-cited problems with L numbers seem to stem repeatedly from deviations from this neat system. I think they screwed up badly when they started using the letters or other descriptors (e.g. L027a, L200 "hi-fin", etc), and lumping fish from different drainages or morphologies under the same number, rather than just assigning new numbers. Synonymising numbers is easy, but splitting one number up into several is harder. Introducing the LDA system was hardly a good idea either.

What lessons can be learnt here? Keep it simple!

[MatsP]

Keep it simple!

That's what the chap leading the project I'm working on keep telling me - and I do agree [just keep making things complicated from time to time].

--
Mats

[Jools]

Guilty as charged. I have been instigating to do away with L Numbers as an identification system for years. They just don't make sense any more and often confuse things horribly.

So, I think the key point here is that we all are in violent agreement they are ingrained common names and the purpose of this thread was not to suggest they were removed. But what about an identification system? Can you explain more about what you see in this? Do you mean we should, for example, change Chaetostoma sp.(L456) to Chaetostoma sp(8) while keeping the common name L456? I thought there was more to it than that?

Just today there were two postings in "What is my catfish" asking about L27. Which L 27? Panaque nigrolineatus? P. cf nigrolineatus? Panaque armbrumsteri? P. cf armbrusteri "Xingu," P. cf armbrusteri "Tocantins"? Wait, maybe they mean L 27 A, L 27 B, or even L 27 C? But wait L 27 C is also L 330 and LDA 63, and L 27 is LDA 77. Really, this is a good system? And L 27 is one of the less complicated ones.

Out of all 460 L-numbers, I think L27 is one of the more complex ones there are more complex ones, but then I see little difference between that position and, say, Mystus carcio, so I don't think it's that important. In what you write above there is only one L-number. There rest are trade names and LDA numbers.

Jools

[Jools]

You're a database guy Jools, you know data should be normalised. The L number is the unique identifier and must never change. All other data, origin, size etc. can and do change so must be kept separate. Some kind of on line catalogue would do the trick.

The online list idea would work, and a re-working of the l-numbers pages could do something really useful with this. Hopefully Shane will post some ideas on what an alternative system might be and I can think about how this might be implemented in this online list.

However, the L-number isn't the unique identifier logically or physically (L200 being a good example of the former) and Planet is keyed on its own species ID. Those sites (which shall remain nameless) that did go down the route of keying on the l-number got themselves into hot water (bad design / layout/ confusion / bad SEO) pretty quickly.

I am breaking my own rule about technical details, but I think the first para of this post speaks to your point and where we might go with this.

Cheers,

Jools

[Shane]

Dude, you are so not up to date

So true

Most of the arguments I am seeing for keeping the L Number system revolve around the fact that they are so widely accepted as common names. I am not trying to change that. I am talking about comming up with a system designed by and for the international aquarium community wherein that community retains control of the system.

L Numbers have built in DRM in the fact that the "holotype" is a copyrighted photograph. This fact has created the vast majority of L Number confusion. I am talking about a system much larger than what L Numbers can do. One that is much more of a "creative commons" in its design and implimentation.

We are currently using several different naming conventions for loricariids in the cat-elog. We have scientific names (Acanthicus adonis) and their variations (Hypostomus cf_ternetzi), L Numbers Spectracanthicus sp(l254)), LDA Numbers (Neoplecostomus sp(LDA101)) species named after a collection locality (Oligancistrus sp`lower_rio_xingu`), made up names (Pseudohemiodon sp`peru_ii`) and PC Numbers (Oxyropsis sp(1)).

The L Number system can't fix this, and if you agree that it can't fix this than you have to agree that the only place for L Numbers is in the "common names" field of a cat-elog entry. A really well designed system should actually work for all siluriformes and we all agree that the simpler the design, the better it would work.

The BIN system would be great, but is not very realistic until we all have home DNA barcoding kits. I like the "most likely genus" and "river name" system, but it is very limiting. We would eventually end up with something like Hypancistrus sp "Xingu 8B" and while this would be an improvement over L Numbers, it could get really complex really fast. There are also many fishes we dont have location information for.

Please keep the ideas, protests, lamentations, etc coming as they are all food for thought.

Still think I am crazy?

-Shane

[Scleropages, posts: 425]

The BIN system would be great, but is not very realistic until we all have home DNA barcoding kits.

It's the 21st century and I still can't find a DNA barcording kit at my local CVS or Walgreen's. What's up with that?

There are also many fishes we dont have location information for.

Good point. Also, what of loricariids in the hobby that were captive bred rather than wild caught? Would it be appropriate to have river/location info in their name? What if the bloodlines for the captive bred specimens came from different rivers, but they are essentially the same species? Are they considered hybrids? Interracial? I understand that we are talking about trying to make a better identification system for naturally occuring species, but these may be some things to consider.

What about species that were introduced into a different river system/habitat by another species (humans, birds, etc.)? Would the P. pardalis caught in Florida have a different listing than the same P. pardalis caught in it's native habitat?

I do like the idea of continuing to use the L# system as it seems to be used quite universally by pleco enthusiasts. I like the idea of adding on suffixes (letters) to distinguish separate species, subspecies, etc. I think it's much easier than creating an entirely new system. However, there a some valid points that have been raised here for not favoring the L# system.

Dude, you are so not up to date This is how it goes, I think.

L027 = Panaque cf. armbrusteri 'Tocantins'
L027a = L330 = Panaque cf. nigrolineatus = Watermelon
L027b = L027c = LDA063 = Panaque cf. armbrusteri 'Xingu' = Platinum
L027c = Panaque cf. armbrusteri (I believe it's cf. nigrolineatus) 'Araguaia' = Goldline
L027c = LDA077 = Panaque armbrusteri = Thunder

And that's just the Brazilian ones (except L330), then theres;

L190 = L330 = Panaque nigrolineatus = the Royal pleco
L191 = Panaque cf suttonorum/nigrolineatus/titan = Dull Eyed/Broken Line
L418 = Panaque titan = Peru Green/Shamper/Shampupa

And then you could go into the Panaque schaeferi/bathyphilus;

L090 = Panaque bathyphilus = Papa Panaque = Small eye
L090a = Panaque bathyphilus = Ojo Chico
L090c = L203 = Panaque schaeferi = Titanic = Big eye
L090e = Panaque bathyphilus = Juanjui

Wow, I think I will bow out from trying to help ID "royal plecos" from now on. In the words of Strongbad: Holy Crap.

[MatsP]

The BIN system would be great, but is not very realistic until we all have home DNA barcoding kits.

It's the 21st century and I still can't find a DNA barcording kit at my local CVS or Walgreen's. What's up with that?
[quote]
There seem to be some people working on this. I remember seeing this one a while back:
http://www.sciencedaily.com/releases/20 ... 120455.htm
and there was another one I've seen recently too.
Here's one that I hadn't seen before:
http://www.forbes.com/sites/tedgreenwal ... e-biotech/
(and the authors name is almost an anagram of Walgreen's! How's that for a coincidence )

--
Mats

[krazyGeoff, posts: 642]

So last night I spent 2 hours typing up a massive response to this very interesting topic, and then on the final preview my browser crashed and all was lost.
It was getting late and I was starting to loose the context of my points, and the post was almost becoming a multi page rambling discussion, so perhaps it was best that it got lost.

The main points from what I remember related to the following:

The L number naming system provides a common language that the most basic of us in the hobby can follow. It's kind of cool to be able to speak the L numbers, makes you feel part of the catfish community.

Perhaps my entire point is this: "The L number naming system provides a common language that the most basic of us in the hobby can follow", and note here that I think of the L numbers as a NAMING system rather than an IDENTIFICATION system.


The scientists amongst us (of which I am not) get more and more frustrated as it appears that the L numbers are taken as gospel, as being accurate and informative, rather than the indicator that they are of potentially a new fish. And as the numbers get bigger it just reminds us all of the amount of scientific reasearch left to do in order to determine whether or not they are a new fish, or just a location based variant (for example).

Just try and find the capture location of a L number in your LFS. I have tried, if you are really lucky the best thing that you can hope for is that someone actually speaks to the importer, and they may be able to provide you the location of where they got the fish sent from, but thats about it!

No-one (so far) has suggested abandoning the L number system.

So what is it that we are trying to do?

Perhaps the time has come to say: "Well the L numbers are a good place to start, and a scientific description is a good place to end, but there is so much good (accurate?) information out there (and on here) that we need a more advanced "language" to categorise these fishes into, because they are most likely never going to be scientifiaclly described."

And perhaps that is all this topic is really about.
We have a common language naming system, a precise scientific identification system, and perhaps the "community" is now maturing to a point that we need a non scientifically ratified identification system?

Another good question is "what is the information used for?" (L number vs. Scientific description)

Personally my journey goes like this:
The LFS ring me and tell me that there are some "new fish on the list", which means the importers have listed some new fish.
Now the LFS isn't going to actually get any (usually), but they will get some if I want them.
So I get the information that they have.
I will end up with a common or trade name, possibly a family name and some kind of genera indicator.
At this point "google is my friend" and I begin the search.
But what am I trying to do? I'm trying to figure out if I can care for this fish. Can I feed it, can I keep water within its "tollerance", is it compatable with existing tank mates, how big might it get.
At this point I become interested in what it might look like, and hopefully I might have an idea of the L number.

BUT this is the extent of what the L number can give me. The L number is most likely wrong, but at least I know whether I'm looking at a 40 cm wood eater or a 9cm omnivore.
One point may be that I know the limitations of the L number system, so its not like I am using this system for identification.

Perhaps the whole point is that if the L number system is used by 80% of the people for the purpose of trying to keep the fish alive and well then it is good enough.

For the other 20% then when all imported fish come with some kind of information relating to capture location and other IDENTIFYING information, then perhaps some other meaningfull system could be devised.
Perhaps I have just rambled on again, but hopefully there will be a few interesting observations above.

[Birger]

This has been discussing L numbers which seems to be the most complicated group so far but if we do happen to come up with some kind of a solution this should really be able to crossover to all "groups" .

C numbers are not far behind L numbers and we are creating our own sort of system right here at PC for unknown, undescribed and hybrid syno's by just "Cat-eLoging" the few that we have which people are now referring to them as (for hybrids this is good as then they are known as just that a hybrid)...but at the moment some have a sp. designation which is not really right as they either have not been scientifically described such as Synodontis sp(7) or are hybrids, so there are people out there referring to their hybrid syno as Synodontis sp_hybrid(1) which they have picked that "name" from here at PC, and we know that most other forum ID's of catfish often contain a link to the Cat-eLog.

Anyway...sounds like we are looking for what people would like to see in a system...those that have seen my talk on synos know that I like to use bioregions that are already in place to divy up the synodontis, I am not sure how to integrate it but I would like to see these become more familiar to people and being used more often as I think it helps people to see the bigger picture of why there does sometimes end up being, as in the case of Auchenoglanis a species or two in each bioregion, or to better understand how a species flock happened to develop in an area.

A quick description of what is a Bioregion? (now in the glossary)

A geographical area described in terms of its unique combination of plants, animals, geology, climate and water features – an area defined by natural boundaries and distinct living communities rather than artificial borders.

Birger