Website founded by
Milan Velimirović
in 2006
13:12 UTC
| |
MatPlus.Net  Forum Internet and Computing MPDBS |
| |
|
|
|
|
You can only view this page!
| | Page: [Previous] [Next] 1 2 3 4 5 | | (61) Posted by Jaro Stun [Friday, Sep 7, 2007 22:11] |
Hello everybody,
my father (Jaroslav Stun - composer from SVK) made me aware of this discussion and after reading some of the posts, I feel I've got something to contribute here as well.
As I understand the problem, main question is whether the developer of database Y (DY) was right to use database X (DX) as a template and whether he was right to re-publish all data from database DX into his database DY.
I see it this way:
Every database can be separated into 2 parts:
1.logical architecture, storage algorithms and methods etc.. => software, and
2.the data itself
Regardless of how much trouble and effort was necessary to put all the problems in CP's database X, the DATA are not his IP. However, all storage algorithms and database architecture ARE.
So,
the best solution for this problem would be for DY to create it's own architecture, tables, relations and all this DB stuff AND a middleware extraction algorithm to convert the data from DX into DY.
On the other side, as far as data are not CP's IP, CP should provide a transparent export mechanism, so the data from DX can be read and published without the need for DX software itself.
DX should just provide a simple XML export template to export problems and nothing more (diagram, and basic problem info [as published in magazines...]), so DY can implement a mechanism to read exported data and convert it into appropriate format for DY.
Then DY can implement his own algorithms for data management, intelligent chess problems sorting algorithms etc.
After this, we've got 2 software solutions that can present the SAME DATA, that are IPs of individual chess composers. Now everybody (DX, DY) can specificate it's own access rules and it's upon the END USER to choose between data access method (DY or DX, or to create own mechanism to read feed from DX's export).
That's my 2 cents,
Jaro Stun [Voniak]
| | | (62) Posted by Harry Fougiaxis [Friday, Sep 7, 2007 23:22] |
QUOTE
- Feist case was decided by The Supreme Court that overruled decisions of lower courts and sided with Feist.
- Since then, a so-called Feist test is used to determine if database can enjoy a (compilation) copyright protection.
- Even if underlying data is not copyrighted, e.g. a collection of facts, database compilation could be copyrighted as soon as it passed the Feist test.
- The Feist test is defined by the Supreme Court as "selected, coordinated, or arranged in such a way that the resulting work as a whole constitutes an original work of authorship"
Any claims how hard it is to create and maintain winchloe database is not relevant argument for US courts anymore. Your database should prove enough originality and pass the Feist test to enjoy copyright protection.
In your opinion, there is not any WinChloe feature listed on this page, which makes this database an original work of authorship?
http://winchloe.free.fr/wc.html
QUOTE
European Union has more strict protection. In particular, no Feist test is required to claim originality.
Directive 96/9/EC of the European Parliament on the legal protection of databases
Chapter II : Copyright
Chapter III : Sui Generis Right
http://europa.eu.int/ISPO/infosoc/legreg/docs/969ec.html
QUOTE
Any claims by Harry that their use of printed or other compilations for Winchloe is not a violation of copyright are not true. Both these uses are subject of the same law. 'Change of media' argument is not relevant under copyright law.
Could you please provide a url where this is written? If it is like that, do I understand correctly that all those who have so far built a database violate some copyright? At least, the author of WinChloe is brave enough to have and keep updated a list of the material that was used.
Note: only 100% recorded and processed sources are listed; in fact, for certain stipulations like proofgames practically whatever has been composed so far is in, fully processed and annotated thanks to the contribution of the tireless Michel Caillaud!
| | | (63) Posted by Vladimir Tyapkin [Saturday, Sep 8, 2007 02:22]; edited by Vladimir Tyapkin [07-09-08] |
For Harry:
QUOTE
In your opinion, there is not any WinChloe feature listed on this page, which makes this database an original work of authorship? http://winchloe.free.fr/wc.html
You know, I am biased. The question of originality is very subjective. It is the same in chess composition: one judge can exclude the problem completely from competition on basis of originality; another may award it the first prize and see it original enough to justify it.
I'll give my opinion anyway.
We should remember that Winchloe package consists of two parts: program and database. This whole discussion is about database part of Winchloe. Most features from your link are available when you have both parts. Let's take, for example, one feature that was mentioned earlier - how Winchloe may automatically describe the problem's theme. If I have just the database part, this feature is not available for me. New problems would not be automatically processed. You need Winchloe program part for that.
To claim originality as a compilation we should find something that is original and part of the database only. Chess problem databases are direct descendants of printed compilations. That's why they are subject of the same law. In printed chess books it is quite common practice to have indexes by year, by composer name, by theme etc. Winchloe database does not add anything new here. If classification by theme never happened before in printed books, e.g. it was Christian’s invention, I would consider it original enough. The idea of classification by some new non trivial criteria would be original. On the other hand, enhancing of known criteria (like themes) is not original enough.
QUOTE
Directive 96/9/EC of the European Parliament on the legal protection of databases
http://europa.eu.int/ISPO/infosoc/legreg/docs/969ec.html
So, what is your interpretation? I am too lazy to read the whole document. In particular: are databases in EU are subject of contract law as in US? The positive answer will resolve the issue quite easily; we read the user agreement that comes with Winchloe and see what is allowed and prohibited to the end user.
QUOTE
Could you please provide a url where this is written?
Well, you provided it for me already: Chapter I, Arcicle I, and Paragraph 2 of your Directive.
"For the purposes of this Directive, 'database' shall mean a collection of independent works, data or other materials arranged in a systematic or methodical way and individually accessible by electronic or other means." Milan's 2345 collection of merediths fits that description. It is collection of problems(that are independent works) accessible with the means of my eyes. I am sure, you would not believe me, but it is quite common interpretation of legal documents.
I was going to find references to US laws, but apparently EU considers it the same way. This part of discussion was where I accused Winchloe contributors of using Minan's work; this directive seems even more relevant because most contributors are in Europe.
QUOTE
If it is like that, do I understand correctly that all those who have so far built a database violate some copyright? At least, the author of WinChloe is brave enough to have and keep updated a list of the material that was used.
Yes, you are getting it correctly. Copyright laws are very ridiculous. There is very little room left to build database legally. The only 100% legal way is to use the original sources only, instead of collections. In our case, only magazines and papers can be used where problem is published originally. You can get some from a collection/database but the amount cannot be significant. The only reason why the world did not fall apart yet – litigation costs a lot of money and only companies with deep pockets can afford it.
QUOTE
Note: only 100% recorded and processed sources are listed; in fact, for certain stipulations like proofgames practically whatever has been composed so far is in, fully processed and annotated thanks to the contribution of the tireless Michel Caillaud!
Well, that means you are admitting more copyright violations. On a serious note, if you point is that the criteria of originality could be the biggest collection in the world for one particular genre (proofgames), I am still disagree. More broad claim is that a pure quantitative factor (amount of problems in the database, the number of unique authors, etc) cannot be a criteria of originality. I have some logical reasoning but too tired to explain it now.
| | | (64) Posted by Sarah Hornecker [Saturday, Sep 8, 2007 11:39]; edited by Sarah Hornecker [07-09-08] |
Ok, one thing about the seemingly high price!
It's 200 Euros für 250.000 problems and studies. Do you always complain that way when Harold van der Heijden brings out a new database (50 Euros for 67k studies) or when somebody writes a book (in the best case 30 Euros for 2000 problems or studies)?
Now, compare the prices. You'll see it is not that much.
You're ripped off if you buy books, not if you buy databases! And as for Mr. Poisson, he also has a solving program etc included in WinChloe. If it is for the database only, I got some information today but want to wait until it is officially announced. Just wait a few days.
| | | (65) Posted by Harry Fougiaxis [Saturday, Sep 8, 2007 14:29] |
QUOTE
If Wincloe user agreement does not prohibit to redistribute database explicitly...,
As far as I remember, there is no EULA when you install the program, there is only a short copyright note in the About screen.
QUOTE
...any US Winchloe user can redistribute the database.
One of the most decent US users, Reb Orrell, preferred instead to process the book 398 Zuglangen Rekorde in Serienzüger by the late Milos Tomasevic, and to write an extensive english guide (on top of the english help file that comes with the program).
http://www.strategems.org/beginners/software/WinChloe/WinChloe.htm
QUOTE
Let's take, for example, one feature that was mentioned earlier - how Winchloe may automatically describe the problem's theme. If I have just the database part, this feature is not available for me.
Pity that this is not exactly what happened. The themes were copied and included in MPDBS as they appear in WinChloe, unless you consider the unsuccessful import of the french accented characters into MPDBS an original feature. The same is valid for a part of the sources (again many of them were transferred to the server in exactly the same french text) and for the solutions (indication of letters in reversal-type directmates, for instance).
QUOTE
Yes, you are getting it correctly. Copyright laws are very ridiculous. There is very little room left to build database legally.
Ok, if it is so then, we should better stick to the ethics/moral/decency aspect of this story.
QUOTE
The only 100% legal way is to use the original sources only, instead of collections. In our case, only magazines and papers can be used where problem is published originally.
Rest assured that WinChloe contributors are doing this too. I could present, if you wish, a list of the thousands of helpmates published in magazines and awarded in formal tourneys, which I entered and annotated into WinChloe.
QUOTE
You can get some from a collection/database but the amount cannot be significant.
In fact, this is what happens with WinChloe. From my own experience (and I mean the books and collections that I processed personally), a good 65-70% of the problems had been already recorded in the database. For these existing entries you have to annotate in their "Reproductions" field your source. Then, and only then, a book appears in the list.
QUOTE
Well, that means you are admitting more copyright violations. On a serious note, if your point is that the criteria of originality could be the biggest collection in the world for one particular genre (proofgames), I still disagree.
I put it in a note, as an indicative example to show that the author and the contributors of WinChloe prefer to work hard, instead of being confused and distracted by doubts whether they infringe copyright. But it can work as an advertising hit, too. Perhaps I should propose it to Christian, so that he improves his marketing which sucks in your opinion.
| | | (66) Posted by Vladimir Tyapkin [Saturday, Sep 8, 2007 19:33] |
QUOTE
...any US Winchloe user can redistribute the database.
One of the most decent US users, Reb Orrell, preferred instead to process the book 398 Zuglangen Rekorde in Serienzüger by the late Milos Tomasevic, and to write an extensive english guide (on top of the english help file that comes with the program).
Don't read it as my appeal to US users to make winchloe available for everybody. It was an example of what legal user could do without copyright infringement, based on my findings. I have a portion of winchloe database, downloaded from its web site. However, I respect your work and have no desire to distribute it. I've recently acquired the database of end games from Harold van der Heijden. I am not going to distribute it as well. But consider, for example you'll get some commercial competition from some tough guy in the USA. Legally, very little you can do to stop him from redistribution of winchloe database. At the very least, Christian should add EULA to winchloe and zip files with some strong language.
QUOTE
Pity that this is not exactly what happened. The themes were copied and included in MPDBS as they appear in WinChloe, unless you consider the unsuccessful import of the french accented characters into MPDBS an original feature. The same is valid for a part of the sources (again many of them were transferred to the server in exactly the same french text) and for the solutions (indication of letters in reversal-type directmates, for instance).
While the themes itself were copied, there is no possibility to process new problems in the same automatic way. My point again - the originality is in program part of winchloe. The database has the end result of it: annotated problem. It is not hard to imagine(and any other chess problem database works the same way), that annotation is done manually. Having just the database part, it is impossible to say was the problem processed automatically or annotated manually.
QUOTE
Ok, if it is so then, we should better stick to the ethics/moral/decency aspect of this story.
I am fully agree. I was saying it from the very beginning. Only after several remarks like 'legally speaking Christian is right', 'Milan should've asked permission', 'the law is on Christian's side', etc., I spent time doing some legal research to prove that it is not as easy as it seems.
QUOTE
Rest assured that WinChloe contributors are doing this too. I could present, if you wish, a list of the thousands of helpmates published in magazines and awarded in formal tourneys, which I entered and annotated into WinChloe.
No need for that. My purpose was to show that copyright laws is a double-edged weapon. If you start a copyright litigation, the best strategy is a counterclaim. It is very difficult these days to do anything without 'stepping into' somebody else's patent or copyright. If Christian asked Milan to remove problems on moral/ethical grounds, I would not say a word. While Milan could refuse at first(I doubt it), public blame and condemnation from members of our community would force him to reconsider.
QUOTE
Perhaps I should propose it to Christian, so that he improves his marketing which sucks in your opinion.
Please don't consider the following paragraph as attack on Christian. I am just trying to give a couple of friendly advices as fellow software developer.
It seems, Christian doing it half for business, half as a hobby. In this case, he is doing everything right. Not care very much about things that are not interesting. If he is serious about making money then my advices could be useful. Trial version is a must, as well as English UI. Online payment method is a big plus. I would consider tiered pricing model for different sets of features. Selling it as a service could work. Hard to tell without seeing hard numbers and sale trends.
| | | (67) Posted by Vladimir Tyapkin [Saturday, Sep 8, 2007 19:56]; edited by Vladimir Tyapkin [07-09-08] |
To Siegfried:
QUOTE
It's 200 Euros für 250.000 problems and studies. Do you always complain that way when Harold van der Heijden brings out a new database (50 Euros for 67k studies) or when somebody writes a book (in the best case 30 Euros for 2000 problems or studies)?
No, I usually don't complain. Besides, it was not a complain, rather a friendly advice. I either buy, or ignore and move on. But you should remember, different people have different values and priorities. I was able to justify Harold's price and I paid 60 Euros(or $82 since dollar is weak agains Euro). I had 2 reasons in mind: 1)support Harold's efforts 2) I like that the database is a regular pgn without any protection.
| | | (68) Posted by Sarah Hornecker [Saturday, Sep 8, 2007 20:26] |
Of course I also could've gotten it someway else but I asked a good friend to order it for me and sent him the money because I want to support Harold. He's doing a great job, working three hours a day on the database. But same goes for Christian Poisson, I like to support him.
I'll send a PM.
| | | (69) Posted by Sarah Hornecker [Sunday, Sep 9, 2007 21:28]; edited by Sarah Hornecker [07-09-09] |
A little bit offtopic, but:
What happened to the White collection? When A.C. White gave it away it contained around 250.000 problems. Is it gone forever or preserved somewhere?
See Schach 1955 (no.06, p.94) here, sadly only german:
http://web.telecom.cz/vaclav.kotesovec/links0.htm
| | | (70) Posted by Vladimir Tyapkin [Sunday, Sep 9, 2007 22:15]; edited by Vladimir Tyapkin [07-09-10] |
It is not gone. Alain White passed it to George Hume. Now it is a property of BCPS. Here is a passage from their catalog:
http://www.bcps.knightsfield.co.uk/libcat.txt
"(a) 23 boxes of the White-Hume collection, mainly direct mates but with some
selfmates. In total, these appear to amount to around 100,000 problems, divided
roughly 5:4:1 between two-movers, three-moves, and longer problems, but the
thematic classification may have caused some duplication."
I read it somewhere in the Problemist, they are ready to loan it for the purpose of digitizing. What is more, they might consider collection to leave Great Britain, that is against the rules of BCPS.
| | | (71) Posted by Sarah Hornecker [Sunday, Sep 9, 2007 22:27]; edited by Sarah Hornecker [07-09-10] |
But then what happened to the other 150.000 problems? Or was the number of 250.000 (by the year 193x) a mistake?
Quotation from Schach, no.06 1955, p.94
QUOTE
Als White sich in den dreißiger Jahren entschloß, seine Sammlung nach England zu geben (...) mögen etwa eine Viertelmillion Probleme vorhanden gewesen sein. Heute ist diese Zahl natürlich bei weitem überschritten
This translates to:
When White decided in the 30s to give his collection to England (...) a quarter of a million problems may have been there. Today this number naturally has been exceeded by far.
| | | (72) Posted by Vladimir Tyapkin [Sunday, Sep 9, 2007 22:36]; edited by Vladimir Tyapkin [07-09-10] |
250.000 number was quoted in multiple sources. I think it was splitted into several smaller collections. It could that other collections from the same catalog are parts of original White collection as well.
We'd better wait till somebody with the access to BCPS library can answer it.
What I find fascinating, is how big this collection was. Only now, computer collections are about the same size.
Update:
Here is an excerpt from an article on Alain White published in October 1942 in Chess Review:
"In 1926, owing to the serious turn in his sister's health, White gave up much of his attention to chess and problems, although the Christmas series continued to appear. His collection of classified problems was shipped to England and there cared for and greatly increased by his friend, the distinguished English composer George Hume, until the latter's death. Care for the White-Hume collection then was assumed by the English expert, C. S. Kipping, who has divided it among a group of assistant curators. In the past year White has resumed the compilation of problem books, through the encouragement of Frank Altschul."
Update 2 on early history of the collection:
This is from obituary written by Comins Mansfield in July 1951 issue of The Problemist:
THE WHITE HUME COLLECTION.
Starting to collect problems in 1892, when he was only 12 years old he began systematic card indexing about 1908 and in 1911 produced some of his results in ' First Steps in Classification of Two-movers.' He stipulated for a standard size of diagram and would send these printed diagrams together, if need arose, with printing sets to an army of copyists, Problems (in more or less their present form, with complete break-away from game positions) had been published since the middle of the last century and there was much lee-way to make up, much research into old magazines and newspapers. Naturally he could not get all the problems which had been published but nevertheless it was a fairly comprehensive collection.
Presumably the Collection began to be housed in their present large boxes which hold some 15,000 problems each. In 1926 the Collection came to England and was administered for ten years by Mr. G. Hume. Thousands of problems were added during this time and in 1936 there were 14 or 15 large boxes.
Others may have collections of upwards of sixty or seventy thousands perhaps (and T.R.D. has an enormous collection of Fairies) but, in spite of the lapse of years, the White-Hume Collection is undoubtedly the fullest and most authoritative up to 1935.
| | | (73) Posted by Michael McDowell [Sunday, Sep 9, 2007 23:53] |
The collection was started by A.C.White and maintained and updated by George Hume. Standardised diagrams would be sent out to composers for them to record their own work. Hume would classify the returns. Towards the end of his life this was a full-time job. The collection was split into different sections (e.g. cross-checks, unpins, etc.) contained in wooden boxes of varying sizes. I don’t think anyone knows how many boxes there were in total. When George Hume died in 1936 C.S.Kipping took over the collection, but given its sheer size he had no option but to farm sections out to various problemists. These have changed hands numerous times over the last 70 years, for example the two-move flights section passed to A.R.Cooper, then R.F.Bradley, then, in 1980, to me. It still resides on top of a cupboard in my mother’s home. A number of the boxes have been in the BCPS library for decades, but the fate of many of the boxes is unknown. In recent years I have occasionally been contacted through the BCPS website by individuals wanting information on a strange box of chess diagrams which has come into their possession, and our librarian John Beasley has been able to recover a number of the boxes (sometimes by donation, sometimes for a fee). Space considerations have meant that John has also had to farm out material, and around a dozen boxes are currently sitting in my garage.
The condition of the boxes and their contents is, in the main, very poor. All have decades of accumulated dust and dirt, and some have severe water damage. Each box contains thousands of diagrams and is a considerable weight. Of course the contents should be transferred to computer, but who has the time and inclination to do it?
| | | (74) Posted by [Monday, Sep 10, 2007 21:02]; edited by [07-09-10] |
Michael McDowell asks>
>The condition of the boxes and their contents is, in the main, very poor. All have decades of accumulated dust and dirt, and some
>have severe water damage. Each box contains thousands of diagrams and is a considerable weight. Of course the contents should be
>transferred to computer, but who has the time and inclination to do it?
It's a question of splitting it up into pieces that can be done a step at a time by several people over several years. The cards (as far as they can take the handling) could be scanned, stored on CD:s or web sites, and then used to transcribe the contents into plain text format at first, with only composer and condition identified, and later more and more detailed markup.
[Added:] There have been some thoughts about digitizing the Toft collection.
| | | (75) Posted by Vladimir Tyapkin [Monday, Sep 10, 2007 21:27]; edited by Vladimir Tyapkin [07-09-11] |
I was going to suggest the same thing. However, even scanning could be very time consuming for hundred thousands of pages. I've done quite a bit of scanning myself and full page scan takes about a minute to produce. It could be reduced with hand-held scanner but not much.
I would consider two other options: to makes photos. Digital cameras have quite good resolution and flash memory is cheap. If place it on tripod and have good source of light, the result would be about the same as scanning for a fraction of time. The second option is to hire professional scanning company.
| | | (76) Posted by Sarah Hornecker [Monday, Sep 10, 2007 21:45]; edited by Sarah Hornecker [07-09-11] |
I sent the same suggestion (scanning) by PM but MMD didn't open it yet. :D
PS, September 11th, 10:33 AM: Now he replied to me
| | | (77) Posted by [Tuesday, Sep 11, 2007 12:25] |
About scanning time: remember that these collections are typically on a fairly standardized card size, usually much smaller than a full sized page. I made some experiments on cards patterned on the Toft collection cards, and came up with a scanning speed of approximately 1000 cards / hour, assuming a Fujitsu fi-4120 scanner, cards approximately 11 x 8 cm in size, scanning resolution 300 dpi and 256 colour scans. There are cheaper scanners out there today: the HP N6010 looks as if it might be useful, for instance (a sheetfed scanner with a straight paper path is probably a must, though). And if the cards are all the same size, of reasonably good quality, and there are no envelopes or paperclipped cards that need manual handling, it would be technically possible to ask a scanning service bureau to scan them with professional equipment. At a cost, yes, but usually very, very fast.
Digital photos could be useful, but good, controlled lighting would be very important. It tends to be something most camera users do not have. Remember, the scans / photos will form a digital archive that should replace the physical one over time: the images should reach a certain archive quality level.
| | | (78) Posted by Harry Fougiaxis [Wednesday, Sep 12, 2007 09:16]; edited by Harry Fougiaxis [07-09-12] |
QUOTE
If you are not a WinChloe user, you did not enter into any agreement with Christian. You are a third party and cannot infringe on Christian's copyright under any circumstances because compilation copyright can be enforced only through contract. That is why Christian should take better care for his database and not allow to download it to everybody.
There is a new version released now. The page with the updates will not include tables for download any more. From now on, the only option will be to get the whole database (ca. 70M) from inside the program.
| | | (79) Posted by Administrator [Wednesday, Sep 12, 2007 16:29] |
Although this is an open and free for all Forum, please abstain from advertizing the commercial software here.
Administrator
| | | (80) Posted by Vladimir Tyapkin [Wednesday, Sep 12, 2007 17:04]; edited by Vladimir Tyapkin [07-09-12] |
Milan, I would not consider it an advertisement, especially in context of the whole discussion.
Besides, it is rather an anti-advertisement. Good for Christian, protecting his database - bad for users, having less options to update. It is a silent agreement with our arguments that the original copyright claim has weak legal merits.
| |
Read more... | Page: [Previous] [Next] 1 2 3 4 5
MatPlus.Net Forum Internet and Computing MPDBS |
|
|
|
ISC 2024
