[Zope3-dev] Re: MessageID's automatic translation in TAL

[Philipp von Weitershausen]

Dmitry Vasiliev wrote:
> Stephan Richter wrote:
> 
>> On Friday 06 May 2005 12:11, Godefroid Chapelle wrote:
>>
>>>>> Stephan also have pointed that automatic translation doesn't allow
>>>>> somebody to disable translation on purpose for some piece of content.
>>>>
>>>>
>>>> Right. That's one reason why it's bad.
>>>
>>>
>>> Can someone explain this use case a bit deeper ?
>>>
>>> If this is a crucial use case, we could still better have
>>> i18n:translate="nothing" or i18n:disable="" or i18n:omit-translate or
>>> whatever on the specific places that needs to disable translation when
>>> the python module author thought it should be translated.
>>
>>
>>
>> Oh come on, this just sucks. I think Philipp raised numerous other 
>> good points I agree with, especially the comparison with Python code. 
>> Really, automatic translation violates the laws of explicitness.
> 
> 
> I think we have more an expected behaviour problem rather an 
> explicitness problem.

Maybe. Though writing code in an i18n-compliant manner is not very 
untuitive most of the time. I've seen good programmers do very simple 
mistakes regarding i18n-aware code. It's ok, it's not a trivial thing. 
IMO, writing i18n-aware code takes experience and with experience, you 
will expect less of the i18n machinery.

> For example when I create a msgid I guess that it 
> will be translated at some point later. For a static content 
> i18n:translate just creates a msgid and then translate it later, why 
> then msgids created in Python code should be explicitly translated? 
> Maybe then we should split i18n:translate to i18n:define and 
> i18n:translate? I don't think so. :-)

The splitting there is YAGNI indeed.

> Sometimes people even don't realize what a msgid is just a definition 
> phase of a translation process, and do things like "str + msgid" and so 
> I guess they will be confused by explicit translation process.

I agree that people often don't realize how to properly deal with 
message ids. The "str + msgid" problem you mention is exactly one of the 
non-intuitive issues I mentioned above.

However, I don't understand your conclusion, I would assume the direct 
opposite. *Because* we have an automatic, magic translation process, 
people don't realize that a message id is just a definition. They must 
think it's the translation process. That assumption is, of course, wrong 
and leads to code like "str + msgid".

> BTW, I've just found another case of automatic translation for evaluated 
> i18nVar values.
> 
> And some more related questions:
> 
> - Should we do automatic translation for msgid's inside a TALES expression?

IMO we shouldn't do automatic translation at all.

> - Maybe we need some new TALES expression type for msgid definition? For 
> example:
> 
>     <tal:block define="name msgid:Specific">

Why? Use case?

Philipp