The Daily Parker

The women's leaders, Ethiopian Ejegayehu Dibaba, 29, and Russian Liliya Shobukhova, 33, run past the 9 km point during today's Chicago Marathon:

7:58 am CDT today, ISO-400, f/5 at 1/400, 55mm, here.

At this writing Shobukhova is in the lead on a 5:17 pace with Dibaba 56 seconds behind her at the 30 km timing pad.

And she has followers:

No, not for the Chicago Marathon, currently underway a city block from me. (Parker and I were cheering the runners on for the past hour.)

Rather, Chicago yesterday set a record for most consecutive days with 100% sunshine since records began in the 1870s:

With 100 percent sunshine today, it marks the 7th straight such day this month - tying the old record set back in October 10-16, 1934. A new October record could be set, if [Sunday], as forecast, turns out to be another day with 100 percent sunshine.

Of course, "Chicago's all-time record for consecutive days with 100 percent sunshine is 10 - set back on July 21-30, 1916," a record we're not likely to break this week as a cold front will hit us Wednesday and return us to autumn.

It doesn't feel like autumn yet, with all that sunshine and a high yesterday of 28°C (and 29°C on Thursday).

That's OK. We'll take it. Parker got two hours of walks yesterday; today he might get three.

Apparently a lot of people are interested in time zones. Here's The Daily Parker's traffic this week:

Fortunately my server seems to be keeping up. I expect that my ISP is unhappy with me, though.

We're having our sixth consecutive day of cloudless skies and warm temperatures. No one's complaining:

Thursday's fifth consecutive 100 percent sunny day was the longest such spell of any here over the past 18 years

There wasn't a cloud in the sky Thursday. It led to Chicago being credited with 100 percent of its possible sunshine for a fifth consecutive day. A check of weather records here reveals it's the first set of five completely sunny days in 18 years.

In an average year, 46 days (13 percent of them) in Chicago record 100 percent of their possible sun. Conversely, 44 days (12 percent) see no sun because of clouds.

The sun and warmth are expected to continue through Wednesday. Why, then, am I sitting inside?

A little housekeeping: if the blog seems slow today, thank this entry, which has got over 70,000 page views yesterday through 19:00 CDT and continues to get hit today. (Usual site traffic is about 4,000 page views per day, total.)

So, there's nothing wrong with either the blog or with your carrier. It's just a lot more traffic than my servers usually get.

After the shocking disappearance of the Olson time zone database yesterday (described here and here), some things have become clearer overnight.

o The wonderful land of Oz has stepped up. Robert Elz, an Australian computer scientist who has actively supported the tzinfo project throughout, has revived the time zone mailing list maintained at the Internet Assigned Numbers Authority (IANA). My, but the list was active overnight, with dozens of people volunteering to host the database, move it to non-U.S. servers, and continue to research and develop it.

o The current database is available from an Australian site at ftp://munnari.oz.au/pub/tzdata2011k.tar.gz. (This is the version running on Weather Now.)

o Astrolabe, Inc., the mom-and-pop Cape Cod outfit responsible for this insanity, has a suddenly-popular Facebook page upon which hundreds of people have expressed themselves. (I wonder if the company will figure out how to disallow wall postings? Oopsi.)

o Software developer Curtis Manwaring, CEO of Zodiasoft Technologies in Las Vegas, claims that he and Astrolabe have had a running fight for years about ownership of various software implementations of time zone data. Says Manwaring:

Gary Christen (CEO of Astrolabe) tried to hire me on three separate occasions and I refused. When Astrolabe obtained the ACS Atlas (when ACS went bankrupt in July 2008), I was concerned that I wouldn't have an atlas for my software anymore. The last time that they tried to hire me, the ACS Atlas became leverage to coerce me into working for them because I had no guarantee that they would give me access to the ACS Atlas. When I refused the last time, I managed to obtain a verbal agreement that I would send my customers to them for a special discount on the ACS Atlas and was relieved for a while. But over the next several months, my customers started complaining about the run around they were getting on the special deal that Gary promised my customers. Eventually one of my customers said that he thought that Astrolabe was trying to make me look bad and disadvantage my business by making it difficult to obtain the stand alone version of the ACS Atlas which is required by my software (but not Solar Fire which has the ACS Atlas bound with it). He was so fed up with the run around Astrolabe gave him that he formatted a database of latitudes and longitudes for me and asked me to add it to my software. That got me back on the issue of researching time zones after which I found the Olson time zone database. I subsequently found from the Olson sources that the ACS data was extremely unreliable, much more than I previously thought.

Note to Curtis, Gary, and everyone else involved in this nonsense: there's a difference between software, which enjoys copyright protection, and data, which does not. This is established, black-letter law in the U.S. (and in most other countries). The fact that both of you produce products that use the same data does not in itself constitute copyright infringement.

What this sounds like—and I'm sorry, Curtis, but you're in it up to your neck—is that you're both amateurs, and your narcissistic dispute has started claiming innocent lives. Arthur Olson and Paul Eggert were trying to help people, as part of the open, collaborative effort we in the software community like to call "open collaboration."

Now, I write software for money, and in fact I have a time zone factory written in .NET, that reads and parses the entire tzinfo database so you can use it in .NET applications. (Send me an email if you want to license it!)

But here's the thing: I know it's a lot less expensive for someone to license my tzinfo parser than to roll their own. Like, two or three orders of magnitude less expensive. And if someone else writes a better parser, they might take my customers away—but that's not copyright infringement (unless they actually use the same C# code or documentation), that's creation.

Speaking of professional software development, I have to start billing now. I'm glad Robert Elz has stepped up, along the dozens of other volunteers, to keep this hidden but vital software project going.

At lunch I thought more about the copyright case against timezone data that the crazy astrologers have launched. I believe Arthur Olson and Paul Eggert, the volunteers who coordinated the tzinfo database for years and who now find themselves sued for doing so, have two principal defenses, one of which may allow them to get the case dismissed.

First, copyright law does not protect strictly-factual information. The Copyright Act only protects the expression of facts. 17 USC 102(b) clearly states:

In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work.

Massachusetts (!) attorney Ronald B. Standler analyzed the copyrightability of factual information in a 2009 essay that relies heavily on the 1991 Supreme Court decision in Feist Publications v Rural Telephone. In the case, a publisher blatantly copied and republished directory listings from a telephone company's white pages. The Supreme Court found that the directory information was not protected by copyright, because it was strictly factual data.

In a later case, Ticketmaster v. Tickets.com, the defendant had used a bot to scrape event information off Ticketmaster's website. The trial court found that the event information was not protected, and the temporary copying required for the bot to operate (it had to download a copy of each page in order to parse it) was fair use.

This brings up the second defense, should the first not win the case for Olson and Eggert. 17 USC 106 allows certain uses of copyrighted works "for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research..." and requires consideration of "the effect of the use upon the potential market for or value of the copyrighted work."

In this case, the tzinfo database (technically a publication of the U.S. government thanks to Olson's employment at the National Institutes of Health) lists historical time zone rules to enable any consumer of the data to find local wall-clock time for any point on earth back to the institution of standard time there. Who cares about this? Well, how about historians? Meteorologists? Frikin' astrologers?

And earlier today I pointed out that, if anything, the Olson database creates a market for the time zone atlas, by referring to it and even providing links to the author.

Oh, and a bonus third defense (which will probably be raised first): the infringement, if any, started waaaay back in the 1990s; 17 USC 507 prohibits a lawsuit "unless it is commenced within three years after the claim accrued." Of course, each time they published a new version of the tzinfo database, it might constitute a new infringement, so I'm not sure about this.

This case stinks. Don't even get me started on the plaintiff's attorney, who's not the cleanest dog in the pound according to the Massachusetts State Ethics Commission. (Molloy "admitted that she had violated the conflict of interest law by appearing before the Sandwich Zoning Board of Appeals on behalf of her private law clients to oppose a special permit application after she had participated in her capacity as a Planning Board member in formulating comments on the same special permit application." Oopsi.)

If this case gets past initial motions I'll be shocked. And as soon as I find out where to send checks, I'll post information about Olson's and Eggert's legal defense funds.

The National Institutes for Health, through a quirk of history, maintainsed the worldwide-standard time zone database until today. A Massachusetts-based company, Astrolabe, Inc., has sued the people who maintain the database for copyright infringement. The company claims to have purchased the rights to The American Atlas, from which the time zone database derived some of its data. From the complaint:

Defendant Olson’s unauthorized reproduction of the Works have been published at ftp://elsie.nci.nih.gov/tzarchive.qz, where the references to historic international time zone data is replete with references to the fact thatthe source for this information is, indeed, the ACS Atlas.

Here are a couple of examples from the database:

# From Paul Eggert (2006-03-22):
# A good source for time zone historical data in the US is
# Thomas G. Shanks, The American Atlas (5th edition),
# San Diego: ACS Publications, Inc. (1991).
# Make sure you have the errata sheet; the book is somewhat useless without it.
# It is the source for most of the pre-1991 US entries below.

Here's an example from the data itself, in the Newfoundland section:

# Rule	NAME	FROM	TO	TYPE	IN	ON	AT	SAVE	LETTER/S
Rule	StJohns	1917	only	-	Apr	 8	2:00	1:00	D
Rule	StJohns	1917	only	-	Sep	17	2:00	0	S
# Whitman gives 1919 Apr 5 and 1920 Apr 5; go with Shanks & Pottenger.

I don't think anyone will deny that Arthur Olson, Paul Eggert, not to mention the hundreds of other people who have maintained the database for years, have used the book in question as a key reference. So here are the questions which, unfortunately, will take the court a couple of years to work out:

1. Is data about when time zone rules changed throughout history protected under copyright?
2. If so, who owns it?
3. If someone owns it, is the Olson database a derivative work under copyright law?
4. If the Olson database does, in fact, derive from the work in question, is it a fair use?
5. Just how stupid are these astrologists, anyway?

Because what you may not know, dear reader, is that almost every Unix-based computer in the world uses this database to set its clock to local time. (All of the applications I've written, starting with Weather Now, use the database as well.) Shutting down the database project will require individual system administrators to update their local copies when rules change. It's not onerous, but it will lead to gaps, particularly in global applications like Weather Now.

What's even stupider about this lawsuit is that comments in the database encourage people to buy the book. So even if Astrolabe owns the copyright to the facts about time zone rules—a troubling proposition—their republication in the Olson database increases the likelihood that they'll make money off it.

Only, facts as such are not protected, so I can't see how Astrolabe can possibly win this suit. I will be contributing to Olson's and Eggert's legal defense fund once they get it set up.

Astrolabe doesn't need to look to the heavens to see how this will turn out.

Update: More here and here.

So says Sullivan, reacting to the news that Sarah Palin is done, one hopes forever:

Sarah Palin said on Wednesday that she will not seek the Republican presidential nomination in 2012, ending months of speculation and leaving the Republican field largely settled.

Palin had left the door open to a run but gave little sign of joining the race to challenge Democratic President Barack Obama. She made it official in a letter to supporters and in an interview with conservative talk radio host Mark Levin.

"After much prayer and serious consideration, I have decided that I will not be seeking the 2012 GOP nomination for president of the United States," she said in the letter.

Sullivan interprets:

Her explanation is, as usual, opaque. But the idea that this person is protecting her family - after putting them all on a reality show, after deploying an infant with Down Syndrome as a book-selling prop, after pushing her son into the military, after sending her elderly dad headfirst into a ravine for a reality TV shot, and after using another young daughter as a campaign press bouncer ... well, it's as ludicrous as almost everything she says. I suspect she knows somewhere that the truth about her will eventually come out in full - as it has already in part, culminating with Joe McGinniss' devastating and exhaustively reported book, The Rogue. And the sheer craziness of this clinically disturbed person would bring it all crashing down. So she's bowing out. Call it cowardice; call it a rare example of sanity; call it a bizarre end to an even weirder game of hide and seek for the past few months. But the bottom line is: we can stop worrying about the threat she posed to this country.

I am not alone in my party to have hoped she'd face the President next November, because I'd like to see him win over 500 electoral votes and 45 or so states (why not 49?). But still, one fewer mentally ill person with national exposure sounds all right to me today.

Update: TPM has a delightful collection of videos reminding us of what we'll miss now Queen Esther is back to Persia.

The Economist has given me a timely help. Their "doing business in" series I've found accurate and useful; I hope this one keeps up the pattern.