Weeabeu Banner

Wikipedia is flawed

Most people see Wikipedia as having the potential of containing the sum of all factual human knowledge. That’s ridiculously impressive. No other site comes close to it in terms of sheer breadth of coverage of information – and depth in many cases also. Having been founded in 2001 though, Wikipedia hasn’t changed its core product much since then and it can be seen in its waning growth.


“EnwikipediagrowthGom” by HenkvD – Own work. Licensed under CC BY-SA 3.0 via Wikimedia Commons – https://commons.wikimedia.org/wiki/File:EnwikipediagrowthGom.PNG#/media/File:EnwikipediagrowthGom.PNG

One might be inclined to think that Wikipedia is running out of things to write about, but it’s a well-known fact that it’s been losing editors faster than it’s been gaining them for several years. In fact, they have several pages discussing the issue. Another problem is the increased rejection rate – Wikipedia has become far more picky in its article acceptance, so as a result less articles are created and edited.

I’m an outsider to Wikipedia and while that means I don’t have the in-depth knowledge that long-term users have, I can approach Wikipedia from the perspective of a potential participant. In this light, I’d like to highlight some of the problems I think Wikipedia has.

Editing is difficult

Whilst writing an informative question/answer on Stack Exchange about the first tsundere character in historical knowledge (for the record Sayaka Yumi from Mazinger Z), I noticed that the Wikipedia entry was wrong. Surprised, I decided to update the information for the benefit of anyone else reading the article in the future.

I knew before editing this that Wikipedia was very exact about how their articles should be written, so I did a few versions locally and then sent off the final copy to the big Wikipedia cloud in the sky. Before this however, I came across an old nemesis – the Wikipedia editor – although now I was much more capable at navigating it than I was many years ago in secondary school.

The first issue with Wikipedia’s editor is that it isn’t WYSIWYG (What You See Is What You Get) – You can’t see your changes in real-time. WYSIWYG editing is ubiquitous in the modern web and while editing plain text blind is fine, doing so with Wikipedia’s complex markup language is really tough. To be fair, there are buttons that generate many of these for you and a preview button, but it’s both hard to parse visually and also makes a smooth editing flow impossible.

Above is WordPress’ editor head. I imagine many people are first introduced to Wikipedia in a similar way to myself – wanting to add some information onto a section, or perhaps create a new page altogether. WordPress’s visual editor would improve both user experience and hence user contribution. In my particular example I just needed to change text, no layout information changed. The best thing about this is that it’s also a WYSIWYG editor, it’s not just a preview live-updating – you’re actually editing the aesthetics to what you see in the input box – And you can always go back to the old-style editor if needs be. The only problem with an editor like this is that it might generate unnecessary layout code, but I think this is a good trade-off as much of this can be eradicated through occasional community editing and smart editor functions.

Citing is pedantic

Wikipedia’s content takes a Verifiability over Truth approach to accepting edits. The theory behind this is that our understanding of things may change over time and Wikipedia chooses to represent the current understanding of human knowledge rather than definiteness. What this means in practice is that entries are strictly non-committal – information is given as opinions or citations from  papers, making sure that it is the source’s fault, rather than Wikipedia’s.

It makes sense to tackle the product this way with the backlash Wikipedia has got from academia as an unreliable source of information, but it has become prohibitive now. I may sound like I’m whining by using my particular case as an example yet again, but it’s a good example of a typical user interaction with the site (and convenient too). My edit was declined because of a lack of citation, which is fair enough, but it seems overly strict to refuse the edit as using the series itself to verify the character type isn’t good enough because it’s open to interpretation. The tsundere character type is simply a character who is hostile towards their romantic interest. There’s not much room for interpretation there.

When I attended the web summit a few years ago, I attended a talk given by Joel Spolsky (from the very popular JoelOnSoftware) about building communities and the problems that come with them. I’ve embedded the video of the talk below (He begins talking about Wikipedia at about 5:15). He mentions the concept of notability of citation, sources are to be reputable if they are to back up information – A paper must be published, Someone famous must claim it in an interview, etc. A prime example of the ludicrous levels the rule has reached is Philip Roth’s correction edits on books that he wrote himself that were rejected for unreliability.

We can’t ask Wikipedia to look for absolute truth as what we know one day may turn out to be false. For many years humans had absolute faith that the world was flat – I’m sure we have some assumptions like that still floating around, and will do for decades to come – perhaps not as large scale though. Verifiability is a sensible choice, it’s attainable, but if Wikipedia wants to resume attracting new contributers, perhaps it should consider easing the requirements – accept a larger range of sources and allow some leeway in subjective information.

Cover Image “Citation needed” by www.futureatlas.com

Print Friendly
Labels: Tech

Anything to add?