Re: [Offline-l] XOWA write up

@Asaf: Thanks for remembering the email! It was indeed the late hours of July 31st [1]. It always feels good to hear from someone who notices your emails. :) @Jon: I had been posting regularly to this mailing list, but was discouraged by the lack of feedback. I've still been releasing XOWA regularly over the past two years, so my enthusiasm for the project hasn't abated.[2] In terms of use cases, XOWA appeals primarily to these types of users: * Technology enthusiasts who are passionate about Wikipedia: XOWA allows a complete, recent copy of Wikipedia on your computer or device. This concept appeals to people who understand the value of Wikipedia and love the idea of having all of it sitting on their computer. XOWA's primary draw here is that it allows you to build a Wikipedia with just the wikitext dump. Of course, some of XOWA's features are particularly distinguishing and attractive in their own right[3]. * Users with limited network availability who want a Wikipedia experience: XOWA reproduces the look and feel of Wikipedia in a browser. This ranges from offering the same page layout (for example, the sidebar on the left) to having similar JavaScript behavior (for example, sortable tables). As such XOWA could be deployed as a drop-in replacement for Wikipedia. Note that XOWA also has an HTTP Server mode, so it can function as an isolated server for a group of local users. Also, XOWA Android has the same look and feel as an older version of the Wikipedia Mobile app (they've since changed that look and feel in the past few months) * MediaWiki users who want an offline version of their wiki: XOWA supports a wide range of MediaWiki functions and extensions. As such, it handles all 820+ known wikis[4] across 250+ languages and 10+ different wiki types. It even handles a read-only Wikidata[5]. It can also be used to offline Wikia wikis, as well as other sites that run on MediaWiki * Developers / boutique users who want to do "custom things" with Wikipedia: XOWA stores all its data in SQLite, does on-demand parsing, and makes this all available through API calls. It's also open-source. As such, various customizations are possible: ** Those who want to generate HTML dumps so they can scrape the HTML for research or analytical purposes ** Those who want to parse any wikitext to HTML to do editing or have their "personal" version of a wiki ** Those who want to build an education-friendly version of Wikipedia without "offensive" content In terms of adoption, XOWA is still very small. XOWA averages something like 600 to 700 downloads per week for the past year [6]. I don't do any self-promotion, and really have been relying on word of mouth. As an independent developer, programming takes up nearly all my time and skill set, and leaves little (if any) for marketing. Of course, I'd love to have any help on this. Just as Asaf indicated, I also know of no direct involvement in XOWA by anyone at Wikimedia, but would welcome any participation. I hope this is enough information for you. Please feel free to let me know if you need anything more here, or reach out to me off-list for further discussion. Thanks! ---- [1]: See https://lists.wikimedia.org/pipermail/offline-l/2016-August/001359.html [2]: For example, here are my releases in 2015: http://xowa.org/home/wiki/Change_log/2015.html [3]: Some of XOWA's notable features: * Up-to date: XOWA imports directly from the dumps at https://dumps.wikimedia.org/backup-index.html . As these dumps are generated on a bi-weekly basis, XOWA allows you to have a recent version of Wikipedia that can be only 2 weeks old. * Complete: XOWA generates every page in Wikipedia, including pages in the Wikipedia namespace ("Did you know" articles), Portal namespace, Category namespace, as well as Template / Module ones. Literally, it provides all of the text content of Wikipedia (and the images as well). * Look and feel: XOWA reproduces as much of the look and feel of Wikipedia as possible. You can see a visual comparison at http://xowa.org/screenshots.html#Comparison * Integrates with other wikis: XOWA navigates seamlessly between the offline wikis. If you're in English Wikipedia, and click on a link for English Wiktionary it will open up the page in your offline English Wiktionary automatically (and if you don't have an offline version, it will open up the online version in a web browser) [4]: Basically, all the wikis at https://dumps.wikimedia.org/backup-index.html [5]: See http://xowa.org/screenshots.html#Wikidata [6]: Keep in mind that XOWA doesn't do any telemetry or analytics, so I'm relying only on download statistics from https://api.github.com/repos/gnosygnu/xowa/releases On Fri, Dec 9, 2016 at 1:20 PM, Asaf Bartov <abartov(a)wikimedia.org> wrote: