Re: Conlang Flag: Voting

Arthaey Angosii wrote:

> I'm not sure if I sounded argumentative in my original post,
A little bit (pedantic is a better word) especially when you were
commenting on such trivial things as the fact that errors are reported
in reverse order.

For my part, I was replying after midnight. It's possible that if I
had been less tired, there would have been sentences that I'd have
noticed a better way of phrasing.

> While I understand your concern over resizing images and getting ugly,
> not-as-originally-intended-to-be-viewed results, I disagree that in
> this particular case it causes substantial harm.
I've thought about it some more and here's what I've decided.

On the voting form, the images are there for only one reason - to show
the voter which code corresponds to which image. Everyone who votes
will have already viewed the images and descriptions on the
conlangflags.htm page. So it's OK to have re-sized images (basically
thumbnails) on the voting form, even if some information is lost,
because the full images are on display elsewhere anyway. So I'll use
re-sized images.

However, I would like the voter to be able to view the full-sized
images as conveniently as possible. What is the HTML code to make the
target of a link (in this case an image) appear in a new window?

The revision I'm working on involves a table of three columns,
containing images of width 240 in cells of width 250. I think that
this format will work well, provided I can let the user view a
full-size version of each image in a new window by clicking on it.

Shanth wrote:

> One more suggestion about the php page wouldn't it be better if similar
> flags (ie. the ones with the same theme) were put in as a single category so
> that we can maybe have a second round of voting to decide upon the final
> form of the flag.
No, it's best for each flag to have a seperate code. Because it's a
preferential voting system, no flag is under any disadvantage or
advantage on account of being similar to another flag. However, once a
winner has been identified, we can always decide afterwards whether we
want to propose variations and have a second round of voting.

Philippe Caquant wrote:

> I don't remember if it has
> been noticed that you can mark, for ex, "equal in
> preference to number 11" without giving any letter as
> number 12. It seems that the program doesn't really
> care, but to me the error should be mentioned to the
> voter.
Sure you can - but this is not an error. The code automatically sets
the "equal in preference" variables to "true" between blank entries,
which is the way it should be.

> Also, I submitted several times as "Tarzan" without
> being rejected. I suppose there will be a human
> intervention at that level ? In case the pseudo is
> perfectly free, let's hope that there won't be two
> people using the same one, otherwise both votes will
> be considered null (?)
If more than one votes is received with the same name, I'll find out
who possesses that name and email them to ask which vote should be
counted. If you use an untraceable pseudonym, you do so at your own
risk.

> Of course, what
> will happen of the votes, and will the result be fair,
> is quite another question.
The author of the condorcet voting calculator I plan to use writes,
"I cannot guarantee the accuracy of these results and recommend they
be computed by hand". Nevertheless, I see no reason not to trust that
it will be accurate.

I'm perfectly willing to publish all the votes (with the voters' names
deleted, of course) if there is an interest in this.

Adrian.

Reply