click for larger view

< ( 6nk/1Qpb2r1/8/1P2p1N1/2P4P/4r3/P5K1/8 w - 0-1 )>

But, as one can see, the end piece of the FEN is bungled. That's enough to throw off Olga (and SCID), as a game viewer needs the extra info in order to know how to play the position.

Here's the right FEN:

<( 6nk/1Qpb2r1/8/1P2p1N1/2P4P/4r3/P5K1/8 w - - 0 1 )>

<Z> Amazing catch! So if I delete this game and then re-submit it with the proper FEN string, it will work fine on <Olga viewer> right?

QUESTION: Why does the so-called "jr. web player" have no problem parsing the crap FEN string? If you switch this game to <pgn4web> viewer it plays perfectly even with a malformed FEN.

Doesn't this mean that <pgn4web> viewer can beat <Olga> in a fight? Doesn't this mean that <Olga> was a piece of crap all along, like <Boomie> said, and has ended its relevancy? Even the creators seem to have abandoned this dog.

================

<MissScarlett> mea culpa if I was indeed the original uploader, though I have no memory of submitting the pgn.

<Z> Just to clarify-

There is only one difference between the crap FEN and the proper FEN you supply, right?

So the only difference is that the proper FEN has an extra dash mark between the last rank notation and the game score notation?

<crap FEN> ...8 w - 0-1

vs

<proper FEN> ...8 w - - 0-1

<<(

click for larger view

)>

White to move, no castling, no ep, 0 on 50-move clock, starting move 1.>

That is to say, FEN's never end with Black winning!

<Missy> I'm going to have to get back to you on that, as I'm in no position on positon.

<Z> Oh ok I think I get it now.

So this is the information I did not include at the end of the FEN:

<White to move, no castling, no ep, 0 on 50-move clock, starting move 1.>

=============

How would I enter that information in a pgn field, exactly?

Second question:

Why is <pgn4web> viewer able to parse and play a FEN in which that information was not included?

<MissScarlett> brilliant! You even gave a link that explains how to do it properly.

On <CGPC>, I vote that you use your current elevated admin duties to parlay yourself into a position of even greater power. You really went the extra mile eh. I can't think of another editor who would throw herself out a window just to get the job done.

Recommended reading for all biographers.

* * * * *

As far as the different behavior of the viewers - this is a common design decision.

Do you try to be fault tolerant and do what ~95% of the cases intended despite the mistakes?

Or do you play 100% strict, and avoid having the ~5% of cases that silently misbehave without any indication?

There's really no absolute correct answer (ok, muddle through with auto-correct while sending up a distress flare maybe).

pgn4web is a *good* viewer, and is intended to be very user-friendly, so I understand their design decision. But they didn't intend to be a DB, and thus could be more lenient.

* * * * *

<MissScarlett>

Well, yes and no. The correct answer is much more scientific than the lazy phrase <effectively the same thing.>

Correct answer using maths:

1. All stubs are fragments, but not all fragments are stubs.

In order to understand these maths you need your A-levels eh. I'm not sure if Staffordshire upon Pillock offered A-levels in maths?

Stub is the generic term, covering an entire taxonomy.

You could also have starting or ending stubs.

You could have just a (tactical) position from a game.

You could blend the above - e.g. if a section (or sections) of a game were missing.

<Z>

<The purest stub would be that which just records the game result.>