Page MenuHomePhabricator

A powerful, handy TemplateTiger
Open, Needs TriagePublic

Description

Currently, TemplateTiger feeds on dumps and is very unhandy, if you are dealing with high-use templates. Therefore please, oh WMF, please provide a tiger which prowls through templates live and which can be piloted as easy as catscan (when it's available).

https://tools.wmflabs.org/templatetiger/

→ «« Man77 »» [de] 18:40, 10 November 2015 (UTC)

This card tracks a proposal from the 2015 Community Wishlist Survey: https://meta.wikimedia.org/wiki/2015_Community_Wishlist_Survey

This proposal received 7 support votes, and was ranked #76 out of 107 proposals. https://meta.wikimedia.org/wiki/2015_Community_Wishlist_Survey/Templates#A_powerful.2C_handy_TemplateTiger

Event Timeline

DannyH created this task.Dec 7 2015, 11:40 PM
DannyH updated the task description. (Show Details)
DannyH raised the priority of this task from to Needs Triage.
DannyH moved this task to Wishlist 51-on on the Community-Wishlist-Survey-2015 board.
DannyH added a subscriber: DannyH.
Restricted Application added subscribers: StudiesWorld, Aklapper. · View Herald TranscriptDec 7 2015, 11:40 PM
IMPORTANT: If you are a community developer interested in working on this task: The Wikimedia Hackathon 2016 (Jerusalem, March 31 - April 3) focuses on #Community-Wishlist-Survey projects. There is some budget for sponsoring volunteer developers. THE DEADLINE TO REQUEST TRAVEL SPONSORSHIP IS TODAY, JANUARY 21. Exceptions can be made for developers focusing on Community Wishlist projects until the end of Sunday 24, but not beyond. If you or someone you know is interested, please REGISTER NOW.
Restricted Application added a subscriber: JEumerus. · View Herald TranscriptJan 21 2016, 2:51 PM
DannyH updated the task description. (Show Details)Feb 6 2016, 12:25 AM
DannyH set Security to None.

Currently this task does not explain very well what exactly is unhandy and not as easy at catscan. The task describes a high-level problem to an underlying "root problem" that still needs to get defined. Knowing the background of the request in this task might provide better insight and lead to better solutions.
Please also see https://secure.phabricator.com/book/phabcontrib/article/describing_problems/ for more information (it is not a Wikimedia site but explains the problem with this task a bit more in detail.)
Thanks!

Aklapper added a subscriber: Man77.Oct 5 2016, 12:25 PM

This task is part of the 2015 Community Wishlist.
As items on the 2015 wishlist can be reproposed for the 2016 Community Wishlist, I would like to point out that in 2016, "Proposals should be discrete, well-defined tasks" (this is a new requirement which will help everybody to have a clearer and shared understanding of what is exactly requested).
So if anybody plans to propose this 2015 wish again for 2016, please strongly consider rewriting this wish by following the 2016 guidelines - essentially, by adding more details, links, and examples to the description, so that editors and developers who are not very familiar with the area can more easily understand it all, and why the feature or bug fix is important, and how they can potentially help.
Thank you a lot!

Man77 added a comment.Oct 30 2016, 6:54 PM

I think that the wish to have a tool which gives you an analysis of the usage of a category is really easy to be understood.

How should it look like? Like petscan, it should have an interface with several options, such as "parameter X is defined", "is not defined", "is defined as Y", "is not defined as Y", "contains Y", or "sort by input in parameter Z" etc. You should be able to logically connect such requirements ("give me a list of articles in xy.wikisource.org, which use template A, in which parameter B is defined as C and sort the result alphabetically by the input in parameter D"). There should not be a low limit of maximum results per page.

Aklapper added a comment.EditedNov 7 2016, 11:07 PM

@Man77: What isn't easy to be understood is which actual underlying problem is being described in this task (and if you have no idea what exactly "TemplateTiger" is and no knowledge what exactly it can do or not do, the current task summary becomes useless).
As far as I understand you would like to search for templates (or categories? or both?) based on some criteria (in order to solve which problem exactly that you're facing when being active on MediaWiki projects?), and that should be expressed in both the task summary and the task description.

"Templatetiger" is the name of an online tool:

It is supposed to be useful for examining the usage of templates. If, for example, a template gets changed or is about to get changed, it is necessary in many cases to check how the template is actually being used: which parameters are used, how are they used, etc.

In order to effectively care for a template, you need to have a "handy" way to get the the relevant, up-to-date data. You need this analysis to see how people use the template in order either adapt the template to its usage or to adapt the usage to the template. If you need to unbreak some kind of mess in a template now, a database from a dump of more than half a year ago is not useful. Period. And even if the current templatetiger were offering you up-to-date information, it is not being user-friendly. I even refuse calling that an interface.

DannyH removed a subscriber: DannyH.Nov 10 2016, 11:44 PM

Thanks. As the words "powerful" and "handy" neither have a clear meaning nor a shared understanding, could you rephrase the task summary to actually describe a problem to solve? Thanks in advance.

"powerful": Can work work with a lot of data and digest them according to the user's need. Can search for results live in the wikis, not in a dump.

"handy": Let me intentionally rephrase it with another term of the same precision: UX. No interface, no visual clues how to rearrange a result, unless you speak the language of the URL like the server. Yet again, topicality is a handy quality.

Problems to be solved: Current tiger only displays up to 30 hits per page. Table is not sortable by mouseclick. An interface for (further) refining or rearranging your search result is necessary. And make it live, or at least almost live. I am really something similar to annoyed that this alone doesn't quality for an obvious problem but for one I repeatedly need to point to.

If we are stealing each other's time, please tell me.

I take I look in my old php-code of templatetiger and must say that it's horrible. There are wild string-construction to generate sql-queries. It's simply not to modify anymore.
I think it could need a complete rewrite in something like Python/Pandas and a complete new UI.

For the records, an item on the 2016 Community Wishlist called "Provide a tool to efficiently analyze the usage of a template" links to this task (which is a description of a goal and a problem, instead of a solution to an unknown problem).
Whether this task actually represents that wishlist item is a very different question (I doubt).