Hacker News Reader

Comment by kevbin | Hacker News Reader

It reduces the overall usefulness and quality of Wikipedia to delete leaves that are referenced from other Wikipedia pages. Nemerle is an example: there are several references to it from other PL pages, that before your "improvement" could be resolved w/in Wikipedia, but now require going outside of Wikipedia to determine that "Nemerle is a high-level statically-typed programming language for the .NET platform."

This kind of "cleaning-up" seems detrimental to wikipedia and, frankly, a waste of your time.

View on HN

It takes me but a few minutes to figure out whether I am going to nominate an article for deletion. A single hit that looks like it might possibly be a reliable source and I'm outta there. In other words, the articles I am proposing for deletion are the lowest of the low hanging fruit. I honestly didn't think anyone would care, and to be honest, most don't -- the only reason anyone even took note of what I'm doing is because I proposed Nemerle for deletion.

I disagree that the clean up is detrimental to Wikipedia. I like to browse Wikipedia for new programming languages, and it is very difficult for me to do so when the "good stuff" is buried in WP's lists and categories. More information is not necessarily better :(

Isn't this actually an argument for a better UI for wikipedia?

Perhaps, but no one has come forth and developed one. But what would such a UI look like? Would it filter based on user-given notability criteria?

Does it need to be at all sophisticated? What's wrong with categorizing lists according to "well-known" and "lesser-known"?

A hyperlink.

I remember, before Wikipedia, there was a big debate on the original C2 Wiki about the usefulness of category pages. There is a fairly widespread opinion that categories are a waste of time, because no ontology can usefully classify the body of information that's out there. While it's natural to want to categorize things, people inevitably use very different systems of categorization (as seen by the dozen or so tags that many Wikipedia pages get), and so they'll never be useful to more than a small subset of people. See also Shirky - "Ontology is Overrated":

http://www.shirky.com/writings/ontology_overrated.html

At the time I left the C2 community, the debate was still raging, but there were still a number of vocal anti-category people. Evidently Wikipedia went the opposite way, but I'd argue that whether it's policy or not, categories are still useless. I always enter Wikipedia via Google; I browse around within it by hyperlinks. The value of Wikipedia is as a store of content, and not as a form of organization.

Counterpoint, I have on occasion found category pages on wikipedia to be enormously useful. I'm not aware of any other way to see in one stroke a huge chunk of specialist vocabulary associated with a single topic. Real-world use case, I was looking for a good name for a project that focused on security. http://en.wikipedia.org/wiki/Category:Medieval_defences is a treasure trove of rather more obscure starting points in a way that http://en.wikipedia.org/wiki/Castle is absolutely not. An alternative use case would be a choice of Java or .NET as a platform, a list of available languages is absolutely helpful in that case.

Category pages are useful as a supplement to good (not-deleted) primary pages.

Categories may be less useful than many people think, but the very fact they are widely used implies there may be a place for them in a wiki, especially an all-encompassing one like wikipedia.

I've found categories very useful; See Alsos and List ofs only go so far. (Most recently, I used them to close a number of predictions about 2010; for example, looking for whether there were any successful terrorist attacks in the US. Hard to google a negative, but looking at a small or empty category is much easier.)

Does your browsing use-case typify how the average user is consuming the wikipedia? I find far often more than I'm hitting something because I searched for it directly (in which case, I find it notable to my own interests automatically since it is what I was explicitly searching for), or indirectly via a link on another page (and if such does not exist, then there is no distraction or loss).

What I find most interesting about this is that you are "cleaning up" Wikipedia so that you can better find new programming languages. You've mentioned this a couple of times.

Do you assume that everyone uses it for the same purpose?

What makes you think that is a valid reason for "cleaning" up the articles?

If I embarked on a "clean-up" project like this and found others' reactions weren't as I predicted, I'd consider that maybe I'm not representative of others; that others may use Wikipedia differently from myself. I would undo what others see as harm and find a less contentious way to improve a shared resource. Rather than trying to fit Wikipedia guidelines (http://en.wikipedia.org/wiki/Wikipedia:Notability) to my own preference (http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion...), I'd consider that maybe my interpretation of those guidelines needs to be re-examined.

In the case of Nemerle, a cursory search of academic and practitioner sites demonstrates it deserves a Wikipedia page. Put it back and go hack on that better wikipedia UI you mentioned.

It's interesting that, presumably in valiant defense of the ready availability of potentially relevant information, HN users have here voted the extremely relevant airing of a not wildly unreasonable argument, directly from the central figure in an internet-wide debate and the personal topic of this thread, down to -4.

I usually dislike and cringe at whinging about downvotes, but I think whatever pattern of thought is at work in such a decision is genuinely relevant to the topic at hand.

>More information is not necessarily better :(

Naivety.

I agree. In the case of wikipedia more information IS better, especially if it's accurate and well written. What kind of person in academia believes that less information, even in this context is a good thing?

> it is very difficult for me to do so when the "good stuff" is buried in WP's lists and categories. More information is not necessarily better :(

Category too crowded? Sure let's just delete a few.

May I suggest you just add a new category for "notable programing languages"?

"several references" is over 50 references

Reply on news.ycombinator.com