I don't mind - I just wanted to get the English sentences matching Japanese ones to avoid characters that aren't compatible with the old JIS codes. Better for searching too.

I don't want to start an edit war, but this sentence used to have a dash, not a hyphen. The "traditional" two hyphens, I think, are used mainly in manuscripts, because the typical old-fashioned manual or electric typewriter has no em dash.

Grammatically, a dash, I believe, is more appropriate for connecting two such disjunct pieces of a sentence. The em dash is now available on Macintosh, PC and Linux/UNIX keyboards (Option-shift-hyphen on a Mac, AltGr-shift-hyphen on Linux and Alt-0151 on a PC). Also, software such as MS Word will by default replace two hyphens with an em dash as part of its Autotext-Autocorrect functionality.

However, if Tatoeba practice is not to use a typographical em dash, I'll go along with that style if desired.

A bit of background, and why I edited the English.

This sentence was one of the ~160,000 Japanese/English sentence pairs from the Tanaka Corpus that were rolled into Tatoeba at the start of the project. They are used in a variety of apps/sites, many of which use the JIS X 0208 coding, which is compatible with ISO8859/ASCII but not with a lot of special characters in Unicode. Several of these systems were operational before the Tatoeba project began, so there are some legacy issues. The conversion of that hyphen to a dash has introduced a character that is not rendering correctly in all the apps/sites.

I was contacted by a user of a down-the-line system and asked if I could fix the ~15 sentences that were having problems. It seemed easier to fix them here, rather than do code interventions on every system using the weekly download of the JE sentence pairs.

There are 128 English sentences in the JE pairs with " - ". This was the only one with a dash.

Thank you, JimBreen, for that very reasonable explanation.