Sunday, 21 January 2007

pne: A picture of a plush toy, halfway between a duck and a platypus, with a green body and a yellow bill and feet. (Default)

I just looked at Amy after breakfast and asked her, “Shall we take off your skirt?” before I realised I had the wrong word, and corrected myself with, “er, your apron?”

I blame interference from German Schürze (which may well be cognate).

pne: A picture of a plush toy, halfway between a duck and a platypus, with a green body and a yellow bill and feet. (Default)

In an article on OpenDocument and Microsoft's Open XML, Bill Poser says that:

Open XML does not follow ISO 8601, the standard for representation of dates and times. Why? Because whoever wrote the code for computing dates in a Microsoft product long ago did not know that 1900 was not a leap year. (Years divisible by 100 are not leap years unless they are divisible by 400.) Open XML requires conforming implementations to replicate this Microsoft bug forever.

On the other hand, I once read a different story which went roughly like this:

  • Excel treats 1900 as a leap year because it's done so since the first version, and can't change that now due to compatibility concerns with people who've come to depend on that behaviour (perhaps because they rely on it for reasons of their own or even just because they have code which specifically works around the behaviour).
  • Excel treated 1900 as a leap year in the first version because Lotus 1-2-3 did so, and they wanted to be as compatible as possible with that software, in order to encourage people to migrate (different behaviour would have meant that some people couldn't transfer over formulas or macros without examining them in detail to see whether the changed behaviour would affect the results).
  • Lotus 1-2-3 treated 1900 as a leap year because it was written for computers with not much memory, and since just using the "divisible by four" rule was good enough for 1901–2099, they just used that.

I may have got the programs wrong; for example, Excel might have inherited the behaviour from VisiCalc, directly or via MultiPlan, rather than from 1-2-3, but at any rate, the story I read is that the behaviour was not specifically planned from scratch at Microsoft.

I don't know whether it's true, but it certainly sounds plausible to me.

But no, let's just assume that Microsoft programmers are incompetent and did not know that 1900 was not a leap year instead. Because after all, we love to hate Microsoft.

(And besides, doesn't ISO 8601 only specify the format in which to output dates and not how to calculate whether a given year is a leap year or not? Does the ISO standard specifically forbid outputting something like "1900-02-29T12:34:56"?)

GetAbstract

Sunday, 21 January 2007 16:19
pne: A picture of a plush toy, halfway between a duck and a platypus, with a green body and a yellow bill and feet. (Default)

A number of years ago, Heise Newsticker linked to a service which provides abstracts of books (mostly business-related ones), so that people can read the most important parts (according to the summariser) of a book in only five to ten pages.

The reason they were mentioned was because of a promotion they were doing: during the time leading up to German elections that year, they provided free abstracts of the main parties' political statements.

I thought that was an interesting service, and downloaded those.

Every now and then (especially around election time), I wonder whether they still do that, but I could never remember the company's name. However, recently another blog linked to a company called GetAbstract, which I believe might have been that company. (They don't seem to have anything appropriate when I search for "Policy & Current Affairs" books written in German, though.)

Still, perhaps that might be useful for other people. (Though you'd have to be willing to shell out several hundred dollars a year for their service.)

pne: A picture of a plush toy, halfway between a duck and a platypus, with a green body and a yellow bill and feet. (Default)

I've seen the "favourite userpic" in three journals so far, slightly different each time :)

Here's how I'll word it:

Comment on this entry, and I will tell you which one of your icons is my favourite. (I may choose more than one if I can't decide which one I like best.) You can comment with the one of mine you like the most, or just with anything.

(Which is a composite of the three I saw.)

As for me, I like to use different userpics in different situations, so it's difficult to pick one favourite of my own. However, I kind of like the 'rgb' icon I'm using on this entry. I suppose it's a geeky kind of thing; at any rate, I can't really put into words why I like it except that the concept seems nifty to me.

Profile

pne: A picture of a plush toy, halfway between a duck and a platypus, with a green body and a yellow bill and feet. (Default)
Philip Newton

June 2015

S M T W T F S
 12 3456
78910111213
14151617181920
2122232425 2627
282930    

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Saturday, 3 January 2026 13:45
Powered by Dreamwidth Studios