<mahmoudkubba: <<To ChG.Com>>: well, can you tell me if the money will be refunded to my card again after the seven days pending passed or what will happen to it? I don't want to lose my money like that. Please help. with respect, mahmoudkubba>

If I recall, mahmoudkubba claimed that his card was charged but our research showed that it was simply declined. I have to assume his funds were taken off hold in short order, because he quickly dropped to topic.

In spite of all of this I'd like to offer mahmoudkubba a "Chessgames Lifetime Achievement Award" of one free year of premium membership.

We don't do this very often but Mahmoud has given this place color since 2004 and I feel sorry for his been locked out in spite of his best efforts.

Don't let this impede your efforts; we need the data and the presentation will be corrected one day.

Somewhere in previous pages you can find Zanzibar and I discussing exactly what tuple* should be used for sorting games. (It's not as simple as the date, since so many games are lacking dates.)

I was just looking into that small request of yours <Could we have an edit button at the top and bottom of the game list?> when I realized that there already is an edit button at the top and bottom of the collection editing page. Therefore I have to assume that you're talking about a different page.

In retrospect I think you might have meant the link (not button) that reads

< » You are the owner of this collection; you have the power to edit it.>

Sure. Here's my thoughts in a nutshell. First, it would be great to know where we procured game scores, if not in the PGN itself at least somewhere in the database. It's somewhat a pity we decide to do this after being around for 14 years but better late than never I suppose.

As you know occasionally you run into a game like H Holmes vs Yates, 1909 that ends with a comment <Source: BCM, Sept 1909, p402>.

Including that information as a closing comment to the game in PGN is clearly not best practices, but we tolerate it when the alternative is throwing the information out. (Now there is another tool: a CG Editor can leave it in the CG Editor notes, but this doesn't do anybody much good since attribution to source is something that should be taken care of when the game enters the database, not as an afterthought for the editors.)

So there is a proposal to add a new database field, call it "source", which would be a string (a VARCHAR in MySQL parlance) which is fed into the database three ways:

(1) Through the same editing template that is used to change dates and rounds etc.

(2) By manually stripping out those ugly closing comments (then GOTO 1)

(3) By including a new ad-hoc field we invent called GameSource that, if spotted on a new upload, would fill in the field properly.

Once all that is done, it's a minor matter to make this new field appear on the HTML of the game page, or even in the PGN itself.

Now, here is my biggest concern by far:

<4.3: Line lengths

PGN data are organized as simple text lines without any special bytes or markers for secondary record structure imposed by specific operating systems. Import format PGN text lines are limited to having a maximum of 255 characters per line including the newline character. Lines with 80 or more printing characters are strongly discouraged because of the difficulties experienced by common text editors with long lines.> From the PGN standard, http://www.saremba.de/chessgml/stan...

In other words, shall we use a VARCHAR(60) or a VARCHAR(40) or what?

The problem is that we are talking about an academic citation, which if properly done can be quite long indeed. That means we are faced with making a compromise one way or another:

(A) Pick a nice small tidy VARCHAR(32) and force the editors to squeeze all citation information into the space. The inevitable pleas for "just a few more characters" will fall on dead ears, because I know that if you give them an inch they'll ask for a mile.

(B) Let the citations be enormous, as academic citations often are, but prohibit them from appearing in the PGN itself.

Well I just looked @ TWIC's PGN for the Polish Championship & they have the clock times mixed in with the game scores. Not sure if that is what you meant by non-Western characters but it is pretty annoying. Looks like they need to be cleaned up better than that to me. I will keep looking to see if I can find these game scores in better shape.

Regardless of who reads them (members, editors, administrators), and of where on the site they may be recited, source citations should be readily intelligible. So, if option <B> be not chosen (resort to full-form, self-explanatory, patently intelligible citations, located outside PGN headers), and option <A> be chosen instead (resort to short-form, likely heavily abbreviated, latently intelligible citations, recited in the source field within PGN headers), then I respectfully submit it is vital to provide a table of standardized abbreviations for sources (doesn't have to conform to any external standard) that any person may consult when reading a citation, in order to enable or to facilitate ready understanding of the citation and its use to check a source (one of the primary purposes of a citation, after all).

Such a table of abbreviations might well prove helpful even in the event option <B> were chosen, for it would provide a hortatory (not compulsory) standard that might be followed by those citing sources. Provision of such a precatory standard might promote consistency and concomitant ease of writing, reading, and using citations, and might help to obviate, or at least to alleviate, ad hoc consternation and potential exaltations of form over substance. Citation, while a discipline in its own right, is in the end but a means used to further the wholesome ends of accuracy, accountability, and verifiabiility -- enduring virtues to be sure, and virtues consistent with the mission here at <cg>.

Finally, I ask respectfully whether options <A> and <B> are, or need be considered as being, mutually exclusive. Perhaps both options might be selected and implemented. If that be reasonable and practicable to accomplish, then I urge the adoption of that conjunctive approach. In any event, I agree that use of source tags, of source citations, or of both should be implemented.

Let me start with a view of one implimentation, a small historical database with about 2k games, of which ~420 have Source tags.

https://zanchess.wordpress.com/2016...

A listing of ~50 long Source tags, which used more than 40 characters.

None, used more than 90 characters.

Let me start with a couple of thought...

1) We want the PGN itself to contain as much information as possible. This is a design goal, and argues for (A), but see 2).

2) If the PGN standard allows for 255 characters, I don't think <CG> needs to provide a more limited hard limit.

Of course, an advisory lower-limit might be a Good Thing, perhaps with a soft warning if exceeded.

And clearly, we want Source tags to be as concise as possible. A future comment should elaborate on that.

* * * * *

I'm not saying you're saying this, by no means, but I want to be clear.

The PGN is no more technical than using a index card in a library. There are common issues of limited space, standardization, references, and cataloging.

<I respectfully submit it is vital to provide a table of standardized abbreviations for sources (doesn't have to conform to any external standard) that any person may consult when reading a citation, in order to enable or to facilitate ready understanding of the citation and its use to check a source (one of the primary purposes of a citation, after all).>

Absolutely correct, this is the area where I think we need full engagement from all, in order to have constructive discussions with the idea of making quick progress.

While avoiding painting ourselves into a corner by adopting bad conventions.

<Such a table of abbreviations might well prove helpful even in the event option <B> were chosen, for it would provide a hortatory (not compulsory) standard that might be followed by those citing sources. Provision of such a precatory standard might promote consistency and concomitant ease of writing, reading, and using citations, and might help to obviate, or at least to alleviate, ad hoc consternation and potential exaltations of form over substance. >

You use a little more elaborate language than I, but yes again. We have examples to borrow from, Harding, Pope, O'Keefe, Hilbert, Davies, Zavaratelli, and (??), all use similar systems of citation.

The chess literature is somewhat limited, and many, many of the sources can easily systematically cataloged, with concise rational abbreviations.

<Finally, I ask respectfully whether options <A> and <B> are, or need be considered as being, mutually exclusive. Perhaps both options might be selected and implemented.>

Yes, fully agreed.

I'm thinking of a hybrid system. Let me give an example of one reference:

<TB G-89 p381/404>

This means TB = Tournament Book, and is intended for a game from a tournament where there is one well-known standard tournament book reference.

G-89 is Game #89 which is found on p381.

The last number gives the PDF page for the digital version of the TB.

If I have the tournament open in SCID, and the TB in acrobat, I can instantly look up the game knowing the above reference.

Notes (for gid updates) cannot be edited directly like opening explorer notes, but currently only by a slip (does the same gid note keep a running history of edited notes, for each gid, or are they separate for each gid note entry). Maybe a system of thinking of how to download or share types of (opening explorer or gid or other) notes could be considered, if the idea is to be able to share note ideas to other authorized editor users if so desired for a particular user.

First, I had been concerned about the line-length issue raised by <chessgames.com> in the above post. Though I had not known the specifics of the limitation (now we all do!), I had always been troubled by the tension between my wanting to provide the fullest citation form possible in a PGN source tag and the need for concision of data-entry therein. I also didn't know quite how to abbreviate or to condense my citations while keeping them self-explanatory. Thanks for explaining more about PGN headers and for analogizing them helpfully to index cards one might find in a card-catalogue in a library.

Second, I definitely appreciate your tb citation form, which includes game number, page reference, and PDF-screen reference in a handy, easy-to-read condensed format. Just by way of example, beyond the tournament-book context, here's what I'm proposing one might see upon implementation of the conjunctive approach to <chessgames.com's> <(A)> and <B>, using H Holmes vs Yates, 1909 a test-case:

<(B)>: Somewhere uutside the PGN, but not after the game-score: "The British Chess Magazine (London: Trubner & Co.), Vol. XXIX, Sept. 1909, Game No. 3,203, 402."

<(A)>: Inside the "source," "Source," or "GameSource" field in the PGN header: "BCM V.29/9.1909 G-3203 402". (Note lack of period to save a character.) "BCM" and the other abbreviations and elements of the format would be spelled out in prefatory generic (for "V.", date-convention, game-number, and pagination) and alphabetically listed specific (e.g. for "BCM") entries in a Table of Abbreviations.

https://zanchess.wordpress.com/2016...

Right now I only have Zavaratelli's bibliography from his Kolisch book. Since I did all the work of transcribing and formatting it for easy reading, that's my prerogative.

Did I mention I better much use all his abbreviations, having arrived at most of them independently already?

I'll add Harding's citation abbreviations from his Blackburne book later.

* * * * *

As usual, I start by asking a few questions:

1) How do you envision using a type-(B) reference?

2) Would a type-(B) reference contain additional information?

3) Excepting for TB references, shouldn't a type-(A) be easily transformed into a type-(B) reference just by expanding the abbreviations?

(Note: TB expansion depends on what tournament a game belongs to... so those refs can also be easily expanded.)

* * * * *

Here's how I'm thinking a type-(A) citation is created and used.

I envision someone like myself reading a periodical or newspaper, playing along with SCID. After inputting all the moves I save the game, getting a dialogue box with the usual canonical 7+1 PGN-tags:

<Event/Site/Date/EventDate/Round/White/Black/Resu- lt>

With SCID, there are also ELO fields, and an ECO field (with a Classify button for auto-filling).

After that, there is a section for <Extra Tags>, which is where the Source citation is added, thusly:

<Source "...">

I would expect all users experienced enough to use such a tag, to know what common abbrevs like BCM, DSZ, CL, and NIC are.

I would also expect an experienced user to appreciate conciseness, they would want to quickly note the source citation, and move on.

So we want to make it quick, easy, and as intuitive as possible.

E.g. I could envision a system where we bulk submit a set of games from a single source, like a periodical.

We might ask <CG> to all us to specific the source just once, at the top of the submission.

<@Source = "BCM v25 (1882)">

Then for each game we might like to use a shorthand for this template:

<[Source "~ G-324 p188/202"]>

Which, after input <CG> processes to become:

<[Source "BCM v25 (1882) G-324 p188/202"]>

Again, the idea is to make it easy on whoever is doing the actual work of inputting.

If someone makes the effort to input a game, it should be a very small increment of additional time to add the Source tag.

It has to be, else we are unlikely to get our volunteers to even use it.

* * * * *

How would the Source tag be used?

Here I can see two obvious uses.

First, one might spot some nonsense in the moves, and want to check the original source to see if the game was properly input. Users with the wherewithal to actually do such a thing, are almost certain to know how to use a type-(A) reference, even if they have to consult an abbreviation lookup.

Secondly, a user might like to read the original source in the hope of finding informative annotations.

Almost all the original literature used for the games had some annotations, however brief or erroneous they might be. Again, users so inclined would find it a small cost to learn how to use concise type-(A) citations.

The only real compelling need for a type-(B) citation, that I can think of presently, is if one were reading a true academic paper on the subject. But then, one would assume such a person would have the expertise as to readily translate between the concise and verbose worlds.

* * * * *

Now, if the source were to be displayed on the game page on <CG>, I would have no qualms about it - with the one exception that it might predispose a new user to use the long citations when submitting PGN.

<Vol. XXIX, Sept. 1909, Game No. 3,203, 402.>

I suppose you're assuming <CG> is using some kind of locale preference for each user?

The generic "3203" would be safer otherwise, as some people use

3,203 and others 3.203 or 3 203 or 3203, etc.

And I also should have my preference for putting a p in front of a page number to help the eye. That's my preference, it might not be accepted practice everywhere. Remember that if I'm working on a bunch of games from a tournament, I'll probably have the source pdf open, and so I just want to keep in on the page number quickly.