Request for comments
Hello, people member of the Wikibase Community User Group or of the Tremendous Wiktionary User Group: @0x010C: @Abbe98: @Addshore: @Amqui: @Andrawaag: @Ariel1024: @Aryamanarora: @Balajijagadesh: @Baruneju: @Benoît Prieur: @Bluerasberry: @Daniel Kinzler (WMDE): @Daniel Mietchen: @Delarouvraie: @Dragan Espenschied: @Epantaleo: @Ernest-Mtl: @GastelEtzwane: @Great11: @JackPotte: @Jberkel: @Jens Ohlig (WMDE): @Jitrixis: @K4-713: @Kimdime: @KuboF Hromoslav: @Kvardek du: @LA2: @LaMèreVeille: @LauraHale: @Loz.ross: @Lydia Pintscher (WMDE): @Lyokoï: @M0tty: @Marcmiquel: @Mardetanha: @Micru: @MParaz: @Nattes à chat: @Nemo bis: @Noé: @Otourly: @Pamputt: @Papuass: @Pharos: @Psychoslave: @Reda Kerbouche: @Rich Farmbrough: @Rodelar: @Rolery02: @Satdeep Gill: @Sebleouf: @Shavtay: @Stalinjeet: @S The Singer: @Susannaanas: @TAKASUGI Shinji: @TaronjaSatsuma: @Teromakotero: @Thibaut120094: @Thiemo Kreuz (WMDE): @Tofeiku: @Tpt: @Trizek (WMF): @VIGNERON: @Vive la Rosière: @Xabier Cañas: @Xenophôn:.
I'm please to announce you the launch of this project. I hope that you will be interested in it, but if it is not the case, please excuse me for this notifiction.
So you are invited to discover the project as it was loozly defined so far. This broad definition is intentional to let you all lattitude to give feedback that will help to shape the project. What was written so far can also of course be improved by your inputs, even the name of the project. So please be bold!
- I didn't received any notification from this page, I just have seen the email. It is great if you want to engage in this project for a long term. My plan was to suggest it to the Community Wishlist Survey 2019 but if you think you can do it, so please go. -- Noé (talk) 06:23, 21 October 2018 (UTC)
- Thank you Noé for the feedback on the lake of notification. I thought that link to the user page was also working, maybe there are additional condition in that case for notification to happen. Here it should be fixed, this time I used the ping template. Please let me know if the ping worked for you this time.
- Making it a community wishlist seems to me to be an excellent idea, and one completely compatible with community involvement. That is, yes I am determined to make this project go ahead on the long term and this come with the aim of engaging the rest of the community right from the start, where of course I count the staff of the Foundation that strive to complete the community wishlist as part of the community. --Psychoslave (talk) 09:06, 21 October 2018 (UTC)
- Where is a link to the project to look at it and play with it? --LauraHale (talk) 11:29, 22 October 2018 (UTC)
Get data and get a project
The group you pinged are highly engaged Wikimedia contributors. Typically none of these people join new community groups, but many of them will contribute to growing communities which already have some projects to demonstrate. The markers that this is a new group is that it has lack of projects to showcase, basic documentation, and early conversation histories.
To get attention and contributors, then find some brave individuals, maybe 2-3 only, who will put in work to do a project and show it off. If you get to that point then the more experienced wiki contributors will start to give something. Of course experienced people will answer questions, but right now, this project needs ongoing time commitments from several people to get started. To find your tireless first colleagues I recommend seeking out newer contributors who are not yet tangled in their own projects. Good luck! This is a good idea for someone to do! Blue Rasberry (talk) 12:04, 21 October 2018 (UTC)
- Thank you @Bluerasberry: for taking time to provide this feedback and for your will to provide guidance. You might be right that none of this people would like to join a new community group. However in doubt I prefer to make people aware of the existence of the initiative and let them chose if they consider that it worths engage or give some feedback. This way if anyone is interested people to engage at any degree at this point they can, and no one will be frustrated for not having been given the opportunity to provide feedback before any really concrete action took place. There is no plane to ping all this people at each change of this project, if this that was your concern. Cheers, Psychoslave (talk) 18:44, 21 October 2018 (UTC)
- Thanks, yes, good plan. Ping me again when this is better established.
- I am anticipating that in the Community Wishlist Survey 2019 this year there will be a Wiktionary proposal so perhaps watch for that and contribute if you think it matches what you want to do. Blue Rasberry (talk) 20:56, 21 October 2018 (UTC)
Hi. I'm not entirely certain the ping here. :) But as this references Wikibase, my question would be would this be a separate install separate from WMF or another install of Wikibase as a separate Wikimedia Foundation project? There are severe potential technical issues on both ends. While a community of potential contributors is important, the more critical element here would be having the necessary developers involved to customize the install, the query engine and to create related applications for use by the new project.
If this is a separate Wikibase install looking to be a user group with WMF funding, very cool. I'd suggest that a really important contact would be developers inside the Mediawiki universe, who also create Wikidata apps. Nothing is plug and play, and we have really big issues with ParaSports Data and our query engine as it incredibly difficult to get it to be free of proprietary Wikidata functions and links. I've finished a Gap Analysis as I am currently in the process of trying to find an institutional partner to overcome some hurdles we have. Let me share that as it may be helpful:
Unless you are having a small Wikibase install, without a query engine that is mostly for the purposes of being human readable, things become really problematic really fast.
If this is for the Wikimedia Foundation, I would focus more on outreach to the staff and explain how such a project would fit into the WMF's Strategic Plan, get involved with that process if it is open to the public, focus more on contacting people inside Wikimedia Deutschland and developers within that organization. Those will be much more critical early on, having very clear support from them, as opposed to having community support.
I would first consider finding two or three developers, (as you need a Wikibase, Mediawiki, Sparql engine install and maintainer, a person who handles the data ontology and upload, and a person who focuses on data output, visualization and query engine. Without those three roles filled, you'll have huge issues.) doing the Wikibase install, creating a user group, attending the conference in Berlin, and after you have this established... then work on having the WMF seek to take it on as a separate project. During this time, I would work hard to make sure that whatever is being developed fits into the Strategy Planning for the WMF and be part of that conversation.
- And did a brief chat with a developer I work with. The reality is, doing this will be really hard no matter what because the underlying Wikibase software is not very easy to use. It has a lot of massive fails. The data compression is not greater. The documentation can be awful or non-existent. So much of the software is proprietary that unless you're doing a standalone Wikibase install designed for human reading, it can be impossible without substantial developer investment time and money wise to get it to work. As the applications appear to be limited, you'd have to draw from people interested in Wikibase generally (of which there are few), people who are in your subject domain with developer skills (of which there may be few) or pay people (who could in theory then be poached by a chapter or the WMF.). The only reason I use Wikibase is out of pure desperation, and its limitations are huge in terms of what we can do without a serious investment of time and money. (php as a language poses problems as it's the lowest paid programming language. so people aren't motivated to gain skills in it. You can end up with really poorly skilled developers as a result as there is little incentive to learn it.) The software is problematic. --LauraHale (talk) 11:26, 22 October 2018 (UTC)
- Hi @LauraHale:, thank you for this abondant input.
- The project aim at having a different Wikibase instance within the Wikimedia infrastructure. Access to this instance from other wikis of Wikimedia is also an aim, although we can delay it.:
- I wasn't aware that Wikidata was relying on non-FLOSS, if you or the developer you a refering to can provide more information or further references about the topic, please do. Also please ping this person here if that is not a problem.
- Regarding used technologies, I'm definitely not interested in relaying on non-FLOSS. Regarding PHP, surely they are a lot of great other technologies out there. But we also have to deal with the existant. On this regard we may look at way to move toward other technologies in the strategy side, but this is really out of scope of this project.
- According to the Wikipedia Signpost, Wikibase incubators are "coming soon", so that might help for this project. @Lydia Pintscher (WMDE): might give us more feedback on this regard.
- Thank you for listing identified roles, that should greatly help to structure the project.
- Cheers, Psychoslave (talk) 02:38, 25 October 2018 (UTC)
- I don't follow Wikibase development like I should because I am more interested in data input, and my developer's cynicism can rub off on me. :) Wikidata's Sixth Birthday says:
- The second one is building out an ecosystem around Wikidata with Wikibase as a core building block of that ecosystem. Since we started the development of Wikidata we always had it in the back of our minds that Wikidata isn’t the only place where Wikibase, the software running Wikidata, will be used. We always imagined other Wikibase instances out there next to Wikidata. However we didn’t have the resources to focus on it and to be honest the rest of the world wasn’t ready for it yet either. This has changed now. Wikidata has grown and more and more institutions, companies and projects are interested in not just contributing to Wikidata but also setting up their own knowledge base to open up their data. So over the last year we got the wheels in motion to create an ecosystem of Wikibase installations around Wikidata. We hope this will open up a lot more data (parts of which can then flow back into Wikidata or be used as references) and take pressure off of Wikidata to host all the data in the world. It’s exciting to see so many different Wikibase instances pop up for a wide range of data. Check out a timeline of them on the Wikibase Registry and read about FactGrid and Rhizome.
- Now let’s take a look at what’s coming. Over the next year the lexicographical data part of Wikidata will grow and power the first exciting and useful applications in that domain. Wikibase will become more and more relevant and usable for running your own knowledge base and thereby help us build an ecosystem around Wikidata and open up more data. On top of that another pretty big change will be the first releases of structured data on Commons. This will not only open up another type of data but will also mean something more fundamental for Wikidata. Our items and properties will be used outside of Wikidata at the core of another project. So far we are mostly conscious of others using our data but not so much our ontology. This is something we will have to be much more aware of when for example creating new properties or changing existing ones. It’s no longer just about the needs of Wikidata but other projects as well. And it doesn’t stop with Commons. With federation features other Wikibase instances will be able to use Wikidata’s items and properties as well in the future. I’m excited about this even if it will probably come with some growing pains. But I’m confident we can handle this ;-) Exciting times are ahead of us. I can’t wait to see what people are going to build with Wikibase, lexicographical data and multimedia data in the coming year.
- So having a Wikibase install more like a farm might be possible in the near future based on future development. It also specifically mentioned lexicographical information as being part of that next big push, which might be useful. I'd go back with more with the advice of the Wiktionary community working more closely with Wikibase developers to make sure needs are met in that framework, and establishing a user group to support that relationship, and serving in an advisory role where possible to explain your needs. Also worth having a look at some of the language tools in development like Ordia.--LauraHale (talk) 16:34, 31 October 2018 (UTC)
What is the meaning of perinially? It is not in any dictionary. Surely you do not mean perineally. The next closest match is perennially, but that doesn’t make much sense in the context. --Lambiam 09:54, 26 October 2018 (UTC) Thank you @Lambiam: for the feedback, the typo has already been rightfully fixed. --Psychoslave (talk) 20:58, 5 November 2018 (UTC)
Structuration de l'information
Je pense que pour structurer les données à partir du Wiktionnaire, il sera nécessaire de s'appuyer sur une structuration telle que présentée récemment dans l'ontologie Ontolex Lexicog Noé (talk) 14:53, 28 February 2019 (UTC)