Want to take part in these discussions? Sign in if you have an account, or apply for one below
Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.
1 to 21 of 21
Following on from the discussion in #9 - #11 here, I have created a tool for finding (and counting) author contributions. As usual, it is a Python script (API in the jargon) compiled to C and then a binary, which is called from Instiki. This time, I have put a separate HTML endpoint on top of the underlyng Python API, whose URL is https://ncatlab.org/nlab/author/Richard+Williamson, with my name replaced by whoever’s. Usual nLab pages are under nlab/show/ rather than nlab/author, the latter is new.
I plan to link to this endpoint from author pages under /show (i.e. usual pages) from the menu, and also at the bottom of every page, where the last named author is given, even if this author does not have their own nLab page (the fact that not every contributor has their own page is the reason for using a different endpoint).
For a little entertainment, try to guess how many pages Urs has contributed to before clicking here!
One interesting point to come out of this, with regard to speeding up page loading, is that Instiki’s mechanism for creating links to pages in the nLab is slow. So that Urs’ page would load in a reasonable time, I created the links by hand rather than use this mechanism.
I plan to link to this endpoint from author pages under /show (i.e. usual pages) from the menu, and also at the bottom of every page, where the last named author is given, even if this author does not have their own nLab page
Now implemented. Feedback and suggestions for improvements welcome. See Richard Williamson for the first case (under ’Pages contributed to’ in the menu), and see computer science at the time of writing for the second case, right at the bottom of the page.
I have a couple of other plans, including presenting on the ’Author - ’ page the most recent 30 (say) contributions an author has made, together with when they made these edits. I’ve begun on this, but am not finished yet; but am done for the day.
This is not only a great contribution Richard, but IMHO also very sensibly thought out, from many perspectives, purposefullness, computational etc.! In fact it is good that the author’s pages, as automatically generated, do not belong to show subdirectory, not only because as you said, not all authors have one, but also that it is good to separate the computer generated from the human generated content, as it differs from many points of view. Whoever wants surely can create a link from her/his basic page to the computer generated author page.
I like my page! It says I have contributed to 1600 pages. Great stuff…. but then one of the tasks I have done in the past was removing spam and thus creating empty pages. A lot of them are listed. Useful but they make the 1600 slightly too large. (I am not complaining!)
Looks like mine is not working, maybe because of my use of a middle initial: link
Re #5: Thanks very much for reporting this, Max. Slightly hilariously, it turns out that Rails cannot by default handle “.” in a URL in its ’routing’ component. I have now fixed this, I believe.
Re #4 and #5: thanks very much for the feedback! I could remove empty pages, Tim, from the list, but removing spam is also useful work, so I’m happy to let them stay :-).
I have a couple of other plans, including presenting on the ’Author - ’ page the most recent 30 (say) contributions an author has made, together with when they made these edits. I’ve begun on this, but am not finished yet; but am done for the day.
Now implemented. See Todd’s page for instance: the section ’Most recent contributions’ is new.
This is nice, but it seems to exacerbate one mistake that casual readers of the Lab often make: to think that the last contributor is the author of the entire page. Even more than a link to information about the last author, I'd really like links to contributions (to go with the links to the human-generated pages that we have always had) in the various edits in the page history. (Either /nlab/revision pages or /nlab/history would do.) That's what's relevant when the last author made only a minor edit.
That’s a good point; it would be nice if the author information at the bottom of the page could be less misleading about that.
Happy to implement whatever is desired. Just let me know concretely how it should look (I could not quite follow exactly what Toby is suggesting in #8).
Just let me know concretely how it should look
I suggest that right after the words
revised on [[date]] by [[name]]
on the bottom of any nLab page, it should have a mentioning of the History-page, maybe like so:
for the list of all authors who contributed to this page see [[page history]]
The pointer to the author-page of the last author which we currently have could maybe come afterwards, only. Or maybe the pointer to it could be shortened and merged with the pointer to the author name.
So I am suggesting to change the bottom line of an entry to something like the following:
revised on [[date]] by [[name]] (see all of [[name's contributions]]). For the list of all authors who have contributed to this entry, see [[history]].
Is it even important for the bottom of the page to mention whoever was the last editor of the page? Why not just say
This page was created collaboratively by many people; for a list of authors who contributed to it with their contributions, see history.
Yes!
Sorry, I realize this was ambiguous. I meant: Yes, that’s a good idea! :-)
I’ve now attempted to implement #12, and also incorporated some ideas from #11 at /history and /revision. Let me know if you’d like to me to tweak something further. The code changes can be seen at github, and more detailed info on the changes can be found in the commit message there.
Looks good, thanks!
Thanks again. Maybe “last revised on…” instead of just “revised on…”?
Re #17: done! Commit can be seen at github.
(As will be seen, this kind of thing is rather trivial; if anyone would like to try it for themselves on future occasions (it is no doubt good for as many as possible to become acquainted with the codebase), just let me know if you need help.)
Thanks for the suggestion, Richard. In principle I’d love to get familiar with the codebase. I don’t have time to actually learn the language right now, but if you keep supplying github links, I’ll try to glance at them and start to acquire a bit of familiarity with the syntax and file structure.
Great! That gives me some motivation to keep supplying the links! I had never used Ruby or Rails before, but am finding that it is not too bad to figure out once one starts delving into it. Ruby especially is very similar to Python, for those who know the latter. Rails is very ’magical’, but once one understands the structure, it is fairly straightforward. I can elaborate if and when you or anybody else has time or interest to look into this. I’d like to lower the barrier for people making contributions, so that people can make small changes concerning things that interest them without having any long-term commitment.
1 to 21 of 21