Deb wrote: I would have supposed that the case hinges on one point only. Is that database copyrighted or not? Astrolabe's argument is that it is and royalties are due to the main developers of it.
Sorry to contradict you on this, Deborah.
To understand the case one has to look at the actual data and one has to make an effort to know and understand them. Without having looked at them, one may speculate, but effectively one cannot contribute to the discussion.
The case does not depend on the fact whether a complete database is copyrighted or not. Olson and Eggert have not used or copied any database.
The have used a few facts from published books. In each case of use, they have weighted and judged them against facts published in other reference works.
The ACS atlas consists to 97% of lists of town names with longitudes and latitudes and of assignments of these towns to a particular timezone history table.
The timezone history tables themselves make up about 3% of the ACS books. Of the timezone history tables, about 99% of the information was taken out of other previous publications. They do not consist of original creation, but of copies form other data.
The tz database by Olson and Eggert does not give ANY assignments of cities to timezone areas. This means they have taken nothing from the main part of the ACS books.
From the timezone history information published in the ACS books Olson and Eggert have taken some facts. They have not taken any of the ACS chosen representation. In fact, Olson has created a far superior representation of timezone history information, which is many times superior to the system of data representation created by ACS.
[side note of me as a computer expert: I do not blame Thomas Shanks, the programmer, for having chosen an inferior system of data representation. Computational power and storage was much more expensive and limited at the time of his work than later, at the time of Olson's work. Olson could use techniques of data representation which were computationally too expensive for Shanks. The maintainers of the ACS database can learn a lot from him, and they should. Olson's work is in the public domain, i.e. free to use for anybody, even without quoting him.]
In the Olson database, on can express facts like
'end daylight saving time on the last Sunday of October at 2 am local time, each year until the end of history'.
In the ACS tables, one would need to know the date of the last sunday in October for each year, and for each country with a different local time, one would have to make a separate table. The ACS tables need twenty or fifty times the space, compared to tzdata, to express the same facts, for this simple rule.
So, Olson and Eggert have used a few data out of a collection which was assembled by ACS out of other publications. Olson and Eggert have created a totally different and much superior representation of these very data. This is within the law, not only in the rather liberal US (in respect to data copyright), but also within the stricter laws of most European countries.
If somebody does not want anyone to use the facts he has collected, he should not PUBLISH them. Facts are not protected by copyright.
The legal case of Astrolabe against Olson and Eggert (and any other user or contributor to the free tzdata project) decays to zero. zero point zero.
Of course, Astrolabe is free to sell the PC Atlas, and to pay its original authors, who created it, some royalties. It is probably obliged to. But this does not mean that it owns the facts, which by their nature are public information.
The courts will show this.