Yes, my concern is not to populate articles with data, because it is
The recent approach of Wikidata to be connected with data repositories and
to harvest data and to feed Wikidata is the best approach.
My concern is connected with the old approach to have a local repository
(probably outdated) to feed wikipedia articles and not only in one
language, but in several languages, because it can work if there is a
community that takes these articles in charge if there is not an automatic
process to updated them.
Personally I consider it a "bad" product and a "bad" architecture. The bots
have been used in the past because there was not a central repository, but
it has been a workaround in my opinion, best or worst it is difficult to
say, that time is was the best because no other possibilities were present,
at the moment it can be considered an old approach, and in my opinion a bad
To make a comparison is like to propose handwritten books when there is the
possibility to have a print machine... I would not imagine the time taken
to update the handwritten books instead of to print a new edition of a
book, it's natural that when the print machine was not present, the hands
were the single tool available and the single solution.
The idea to invite people to edit with the sample justification that the
presence of a stub invites to edit doesn't justify the approach because
Cebuan and Waray Waray have not increased active editors since 2014.
The single change happened is only more presence of those language in the
web, useful or not useful to the web crawlers is another question.
On Mon, Jul 6, 2015 at 1:00 PM, Gerard Meijssen <gerard.meijs...@gmail.com>
> Given that this is the Wikimedia mailing list, the assumption that
> Wikipedia is primary is not necessary. The objective of the Wikimedia
> Foundation is in this more relevant. Consequently the balance for an
> argument is different.
> I blog often about issues with Wikidata and Wikipedia. I often inform about
> approaches that make Wikidata more relevant and I am quite happy with the
> progress that has been made.
> However, Wikipedia think does harm Wikidata.
> On 6 July 2015 at 12:40, Marco Chiesa <chiesa.ma...@gmail.com> wrote:
> > On Mon, Jul 6, 2015 at 11:53 AM, Gerard Meijssen
> > <gerard.meijs...@gmail.com> wrote:
> > > Hoi,
> > > When people do not edit Wikipedia we are talking about a situation that
> > > does not exist.
> > >
> > > I wholeheartedly disagree with you as you mistake the process with the
> > > product. Our aim is to produce a product and we should endeavour to
> > provide
> > > it in a SMART way. We lose out when we do not do the best we can. Our
> > best
> > > is NOT sitting on our hands keeping information that is available to
> > > ourselves for secondary reasons. Yes, we can write articles and yes
> > > may be better but as long as we do not provide information we do a half
> > > arsed job. A job that is qualitatively and quantitatively inferior.
> > > Thanks,
> > This is a kind of discussion that cyclically comes back, and I guess
> > there is no correct answer. Wikipedia was born on the internet, and is
> > designed to work on the internet, which means that as soon as it
> > provides good links and good connection between links, it helps to get
> > access to knowledge. It is not something that is supposed to be read
> > from the beginning to the end (it's simply too big), so it is
> > difficult to talk about the quality of the product as a whole. There
> > are good articles, bad articles, wrong articles, uncovered topics; so
> > the experience really depends on the reader's needs. Automatically
> > created articles generally offer a good base, they're as accurate as
> > their sources in providing basic data, and are often a good base to
> > build upon. In some cases no one will build upon them, but even thus
> > they do fulfill a need. Indicators are a simplification of a more
> > complex object, so they can only tell a limited amount of things. We
> > know that article count can be inflated by automatically creating
> > stubs on very specialistic topics, average page weight by adding
> > code-rich templates and so on, but as soon as people are aware of the
> > fact, I don't see the problem.
> > We are now at a stage where a huge number of articles already exist,
> > and where a big share of the creation content can be automated; if we
> > stop now building the encyclopedia, we still have a great product that
> > can help people; a lot of data could be kept up to date using only
> > automated tools, although it will eventually look "old". At the end of
> > the day, classical Latin literature has had no active community for
> > centuries, and is still perfectly usable.
> > Marco (Cruccone)
> > _______________________________________________
> > Wikimedia-l mailing list, guidelines at:
> > https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines
> > Wikimediafirstname.lastname@example.org
> > Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
> > <mailto:wikimedia-l-requ...@lists.wikimedia.org?subject=unsubscribe>
> Wikimedia-l mailing list, guidelines at:
> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
Verein zur Förderung Freien Wissens
Association pour l’avancement des connaissances libre
Associazione per il sostegno alla conoscenza libera
Switzerland - 8008 Zürich
Wikipedia: Ilario <https://meta.wikimedia.org/wiki/User:Ilario>
Facebook: Ilario Valdelli <https://www.facebook.com/ivaldelli>
Twitter: Ilario Valdelli <https://twitter.com/ilariovaldelli>
Linkedin: Ilario Valdelli <http://www.linkedin.com/profile/view?id=6724469>
Wikimedia-l mailing list, guidelines at: