[josm-dev] Change to changeset comment handling, RfD

[Ulf Lamping]

Am 31.07.2010 21:23, schrieb Frederik Ramm:
> For a while now, JOSM has forced people to add a changeset comment that
> is at least three characters long, and refused to perform an upload
> otherwise. In addition, JOSM *always* loaded that last recently used
> changeset comment as a default, making it easy for people to erroneously
> continue using an old comment.

IMHO this is worse than have no comment at all. I occasionally do that 
mistake myself.

> * The last recently used changeset description is only loaded as default
> if it was used less than 4 hours ago. Otherwise there is no default.

I don't think having a default here is a good idea at all. People seem 
to errorneously upload changesets with a wrong (old) description, as it 
was already filled in. I don't think the 4 hours will help, as this 
still keeps the problem of subsequent changes with a wrong comment.

In fact the 4 hours will confuse people, as they don't really get the 
idea what's going on ("sometimes JOSM asks me for a comment, sometimes 
not"?!?).

Why not simply keep the field empty and having the recent entries in the 
drop down - its sufficient enough IMHO.

> * If you try to upload a changeset with a comment of less than 10 (!)
> characters, there's a dialog that explains the importance of changeset
> comments, and asks the user to reconsider.

I don't think that the change to 10 chars is a good idea. E.g. if I only 
add a tag to a node, the comment "tag added" is IMO sufficient but still 
"rings the bell".

Next release we'll might even get: You have to enter at least an upper 
and lower case char, a digit ... ;-)


What I really don't like about that dialog: It tells you that comments 
are soooo useful and that soooo many mappers will look at the changes 
(which BTW in a lot of places around the world is probably misleading). 
But it don't tell the user how a meaningful comment or some good/bad 
examples look like.

The text "This is technically allowed" is again misleading. Although 
valueable and helpful, setting a comment is simply optional. Please keep 
your personal beliefs (that are not generally backed up by the 
community) out of a software that is used by a lot of mappers.


[and from another mail in this thread]
> I think if something is as important as useful changeset
> comments,

Sorry, you overemphasize the importance of changeset comments.

> then it is ok to not make it *too* easy for users to misbehave.

Please keep in mind, that a changeset comment is *optional*, and there 
is no "misbehaviour" if not added.

If a user don't want to add comments it is his right to do so.

Regards, ULFL

P.S: Next, someone gets annoyed by JOSMs behaviour so much, that he 
writes the "Fuck you, Frederik!" plugin, so the plugin will enter this 
sentence to the comment field automatically. I don't think that's the 
way we should end with this ...