2016年5月15日日曜日、Kato, Tomoyuki<kato.tomoyuki@jp.fujitsu.com>さんは書きました:

> >> Hmm... It seems a problem is limited to "stable-mitaka vs stable/mitaka".
> >>
> >> It seems 'master' version of translation stats also has a problem.
> >
> > okay, good information. thanks.
> >
> >> I translated more than 2000 words in 'master' version, but
> >> stackalytics says I translated only 251 words.
> >
> > To understand the situation more exactly,
> > which do you mean 2000 "Words" or "Messages" at Zanata?
>
> What is not clear?
> As I wrote above, I translated more than 2000 'words'.

IIRC, Zanata returns "Messages" (i.e. units, not "Words") counts as stats
from API 'stats/user/%s/%s..%s' % (zanata_user_id,start_date, end_date))'.

To clarify the situation more, my statistics from Zanata is based on

Zanata personal dashboard. That is all what I can know as an usual user.

On the other hand, Stackalitcs uses "terms".

In my understandng, "terms" means "words".

Thanks,

Akihiro

Carlos, Alex, is it correct?

> >> Is it better that I stop translation?
> >
> > If exact stats is important for you, you should stop translation.
> > Otherwise, translation process itself works fine.
> > Also, I'm not sure, but we will get proper stats
> > after stackalytics works fine.
>
> It is an irony :-)
>
> Once translation statistics is available publicly, the 'number' should
> be correct.
> Do you plan to determine ATC status based on Stackalytics stats?
> If so, I think it is a problem.
> My e-mail just wants to warn you all.

Yes. Good finding.
We need to investigate and clarify it.

KATO Tomoyuki

_______________________________________________
OpenStack-I18n mailing list
OpenStack-I18n@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-i18n