Quote:
Originally Posted by highwayman274
There has been a lot of discussion in this section about the reliability of the model names published at different index sites.
|
Yes. @highwayman I'm quoting you here but only because it was your above post/phrase that triggered the following (stray?) train of thought.
I'm in no way criticising what you decided in the above instance.
I've not listed any sources in this thread and that's with some reason. I think the search for
reliable sources is, underneath, an attempted short-cut. The search for short-cuts is a good thing,
e.g. the world needs truly lazy programmers. But we need to be aware of their occasional pitfalls.
There are plenty of very good/highly-reliable sources... no need to list here, it's just that I don't think that's the right thinking or approach. Any more than we should rely upon another MIR poster's repute: do they make lots or errors in which case their next suggestion is likely to be wrong? No, that's a very dangerous path. Do they make no errors so their next suggestion will surely be correct? Equally dangerous/mistaken.
I think the thing that counts is evidence, i.e. academic-style citation. No, I'm not saying we need to start adding footnotes to our posts like wikip., but MIR generally requires links to show evidence and that's very much along similar lines/thinking.
I think there are e.g. websites with highly suspect information, some of which we already know about or can find examples of, but as things move towards the other end of the (imaginary?) reliability "spectrum" they become very difficult/troublesome. Both iafd & egafd have erroneous info., yet they're still generally highly reliable with good reason. How do we find the errors so we can avoid them? By finding and using evidence. xyz is named abc on iafd but they're wrong as shown here at... yadda, yadda.
I think it's a great idea to separate MIR from the rest of vef as a section for "dealing with" id issues. At the same time I think the reluctance of mods from other sections to countenance further "amendments" to existing threads (e.g. changes of name) is troubling. No, I'm not saying I think "discussion" should occur in model threads but I am saying, for example, there should be less (apparent) resistance to changes to model thread titles when evidence is produced.
Oh, so I'm just having another grizzle about mods from other sections? No, that's not my motivation or purpose. I'm trying to look at a wider picture.
Edit: What I was trying to get at immediately above is: sometimes MIR feels like it's pressured to always find a perfect id. OK, understandable because it's best to start out on the right track, cheaper to fix errors early, but I think we should also recognise when sometimes we've possibly reached all we're going to get using the MIR method. This is software dev. too: the construction metaphor is often false/misleading. Often more useful is the gardening one of iterative re-work. That is sometimes the case with model ids, too. MIR may not always have produced a perfect id but after a fair bit of further time that may be the best we can manage, and then if we've got something usable for a handle it may be time to try another method--e.g. model thread--and hope that it can be improved upon, i.e. consciously leaving the door ajar for such improvement.
Sounds like I'm lowering the standard of evidence on model ids? I don't think so. We must keep it as high as possible for as many as possible. Just recognise that there's sometimes other approaches required.
Also just to be clear: my above comments are in no way a disagreement about, for example, beutelwolf's comments about the reliability or otherwise of certain 1960s/70s mags, etc.