Nepomuk Metadata Extractor moved to KDE Review

[Sebastian Kügler]

Hi,

On Sunday, December 02, 2012 21:05:08 Jörg Ehrichs wrote:
> So it's been a while and I've changed "most" part of what was
> suggested. (thanks for the hints again)
> 
> The program changed its name to "Nepomuk-WebMiner" for better
> indication what it actually does and it respects the users privacy as
> well as not to bog the system to much in the default behaviour. Also
> all pre-generated headers are removed now.
> 
> As Sebastian mentioned in his last mail, a few things are left in
> order to get this into KDE SC.
> But as I aim for Extragear at the moment, I'll leave them for now.
> 
> My plan for the future:
> * rename the git repo to nepomuk-webminer (together with the move to
> extragear)
> 
> * figure out how to integrate all parts better into nepomuk core for 4.11
> ** I will "remove" my resource extarctors and add the last missing
> parts (pdf/odf extraction, filename stuff) to the nepomuk-core
> fileindexer
> ** Integrate the fetching status into the nepomuk systray
> ** Combine the KCM with the Nepomuk's kcm (as far as I remember
> vishesh wanted to change this anyway)
> ** And check how the webextractors can be integrated better.
> ** and possible more...

This sounds a bit like you're ignoring most of the feedback you got. That's 
not quite the point of a review. Reviews are not just for KDE SC, but for code 
officially released under the KDE umbrella. QUality of that code affects the 
reputation of all of KDE's products, so that's clearly a problem in my books.

> All this is the plan for 4.11, but I haven't looked into it to check
> how exactly I like to do all of this.
> 
> So in case no one has any objections,
> I will ask the sysadmins next weekend to move this into Extragear/base

One more thing I ran into when installing it (already three weeks ago) is that 
it uses a lot of dependencies, many of them python libraries to interact with 
specific webservices. Many of them are not mentioned specifically by CMake, as 
their runtime dependencies, so it took me quite some time to figure out those 
dependencies. This will most likely lead to packagers having problems getting 
nepomuk-webminer packaged in a fashion that it actually works.

I also see this as a problem for a possible KDE SC integration. Without a lot 
of external dependencies, the project is fairly useless. I'm also not sure we 
can sensibly depend on PyKDE in modules as central as kde-runtime or 
workspace.

Do you see a way to a) reduce the number of dependencies or at least make it 
more obvious how to install nepomuk-webminer in a more obvious way?

Cheers,
-- 
sebas

http://www.kde.org | http://vizZzion.org | GPG Key ID: 9119 0EF9