Hi Jody,
in principle, I agree. And as a maven user, I don't have a particular 
problem with it, having javadocs attached etc. and using geotools regularly.

But I still think that there is a huge amount of people that uses 
geotools as a dependency, not as their primary tool of trade nor main 
interest. I have heard this particular sort of frustration voiced around 
a fair bit. These people may download a copy of the necessary jars once, 
compile, and leave until a new feature is required. You do not want to 
update your existing codebase then.That's hwy I think the release 
javadocs should be available, and the default link should be to this branch.

Also, I would assume that the javadocs are released with the 
source/binaries, not filled in post release and only based on feedback. 
That should be merely used as way of improvement.
When starting with geotools a couple of years ago I found it daunting to 
use features if they are not documented, especially with the recent 
dramatic changes in some interfaces. I think the how to and the wiki are 
a great step ahead, but I have troubles with it focus on snapshot 
release features, instead of latest stable one. Maybe we need to clearly 
indicate in the how-to new and and stable articles...

I think the community could be made more involved in writing these 
how-tos (and I am ready to help), but a clear indication how to proceed 
and what are the boundaries should be given from the main developers. 
Personally, I would prefer a sort of senior-peer review approach, to 
feel comfortable that what is contributed is actually a good example and 
not a dirty hack ;)

Cheers
Martin


On 5/7/2010 9:59 AM, Jody Garnett wrote:
> Hey Martin thanks for the feedback.
>
> With respect to most recent javadocs / intros etc feedback is the benefit the 
> user list provides the developers. Sheltering on the stable branch does not 
> offer the same reward. In short if you want developers to write docs we are 
> only going to be able to motivate them to write docs for the most recent 
> version.
>
> That said we should be able to make the old docs available for download; I 
> will experiment with making a doc download when I release GeoTools next week.
>
> For repositories it is a bit more tricky. Part of the thing is that it is 
> impolite to turn repositories off - since older projects will still build 
> against them and we get very much yelled at for breaking other people's 
> builds. The other complicating factor is that the "stable" repository hosted 
> at osgeo.org has not been stable in terms of the server being up.
>
> Still I think there is a good solution: get geotools into root maven 
> repositories; geoserver manages to get in because they publish to codehaus 
> which synchronises with the root maven repositories.
>
> This way we would only advertise one repository for SNAPSHOTs and advise 
> people to use their local maven mirror for stable releases.
>
> Jody
>
> On 07/05/2010, at 5:10 PM, Martin Tomko wrote:
>
>    
>> Hi Jodi, all,
>> I just noticed this thread and wanted to comment.
>>
>> First, I think that the default javadoc presented should be the latest
>> stable branch.My understanding is that 2.7 is still not a stable
>> release, and I would prefer not to be lured into it until stable, and I
>> assume others may feel similar. It confuses ordinary users that do not
>> follow the development on a daily basis.
>>
>> I would also prefer if the version number appeared in the url, it would
>> make it easy to navigate between  versions, somethign i used quite a few
>> times with other projects to compare.
>>
>> Finally, there seems to be a huge number of various geotools
>> repositories, and if one only searches for it on the web and does not
>> naviagte from the geotools.org website, it is easy to end up there,
>> without knowing how recent the javadocs are. These are the ones hosted
>> on codehouse, on http://javadoc.geotools.fr/, and similar. If any of
>> these are within reach of the community and not actively synchronized, I
>> would advocate removing them.
>>
>> My 2 cents,
>> M.
>>
>> ------------------------------------------------------------------------------
>>
>> _______________________________________________
>> Geotools-gt2-users mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/geotools-gt2-users
>>      
>
>
>    


-- 
Martin Tomko
Postdoctoral Research Assistant

Geographic Information Systems Division
Department of Geography
University of Zurich - Irchel
Winterthurerstr. 190
CH-8057 Zurich, Switzerland

email:  [email protected]
site:   http://www.geo.uzh.ch/~mtomko
mob:    +41-788 629 558
tel:    +41-44-6355256
fax:    +41-44-6356848


------------------------------------------------------------------------------

_______________________________________________
Geotools-gt2-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-gt2-users

Reply via email to