- You visit the main Articles & Resources page and try to drill down through categories to find the article(s) listed under a certain category. You quickly realize this is next to impossible.
- You try and search for the article in the Resources section and it comes up empty, even though you know that article is supposed to exist here.
- You try #2 and actually get results, but those results happen to be an entire article per result, so you can potentially have to scroll through whole articles to view all the search results if the one you are looking for wasn't the first.
Problem #1: Correct me if I'm wrong, but most of the time when you want to find an article after a certain topic or title your first instinct is to run a search for it, right? You may consider looking for the category the article(s) would most likely reside under to see what we have available. Then you realize how many categories we have, and there is a chance your topic could fall under multiple ones so you would need to check them all. Then you need to drill down to that category to see the list of articles, possibly page through them if there are a lot in that category... point is it's always easiest just to search. So we're going to do away with deep categorization of articles (as part of a larger, site-wide categorization scheme) and rely solely on the search to let you find all that we have available for a certain topic or title.
Deep categories (like Programming->AI or Programming->Graphics APIs->DirectX) will still exist for resources, which are links to files and websites out there on the internet, and are more commonly found in a listing format (like blog rolls) than through a search.
Problem #2: So if we're going to rely on a search so much we need to fix it right? Right! There are some articles stuck in a sort of search limbo thanks to the way they have been categorized in the system so that when you try to search for them, the system responds with no results, but the article really is there. If you happen to track down its URL in the old system and try bringing it over to the new one you could wind up seeing this page. So you still can't even access it. Ack! Luckily this is an easy fix of moving the article in question out of the offending "general" category it got placed in originally and into a specific one. As they are found they will be brought back into the light of search and accessibility.
Problem #3: Obviously the solution here is to fix the search template, and just as obviously it's an item on the ToDo list. We will most likely leave out any of the article content in the search results, although I guess I understand why they were there in the first place - to let you see what results were actually relevant to you by looking at the resource content - but a lot of our content is just too long for that to be feasible. We will most likely replace it instead with the "teaser" field of text, as its purpose is to summarize the content. Seems much more reasonable (and short!).
Now that we've covered the system, let's have a look at the articles themselves.
So there have been plenty of complaints about the legacy articles having no formatting and one of my tasks is to fix that - mainly by hand unfortunately. The WYSIWIG editor does make this easier thankfully, at least I'm no longer formatting HTML in a text editor while building articles. Not all articles have come across looking this bad, but even in ones where the formatting is simple and thus unaffected by the switchover they can also be missing images. Some of the articles have downloadable files attached, and these were not carried over (one of my recent additions was the ability for article templates to show attached files, but they still need to be added to more articles that are missing them). There are a few that are so long they even get cut off and the article system won't show any additional text - this is a known issue we're working to resolve. Pagination of articles has some bugs in it, and the print version of articles needs some serious formatting work, as currently it just strips styling from the page and dumps it back into your browser as plain text - the entire page.
As you can see from the previous article link though, when things get cleaned up they can look rather nice. Another example of improvement can be seen here, although you may notice close to the end that I'm still trying to get tables to behave properly.
Oh and article comments still seem to be missing, or the comment count is incorrect.
I'm pointing out all these issues so that you know that we're aware of them, and also so that we make sure you're aware of them, and can help us out by notifying us of them when you see them and where - specifically bad content formatting, missing images, missing support files or bad article URLs. Best way of doing that is by using the Feedback tab you see in the lower-left side of your screen. Obviously I'll be getting around to all the articles in due time, but if I see one pop up in the issues tracker I'll hop on it first.
As a bonus though, in my formatting journeys so far I came across the annoying ability of the site to transform b) into an emoticon - and you can't turn them off for articles like you can for forum posts. But wait - how was I able to type b)?? Well I was told you can change the text that triggers emoticons in the system, so I changed it from b) to "cool" (replace quotes with a colons). I'm really sorry that my original "fix" for this issue was "disable emoticon use" and wish I had known about this earlier!
Now if you'll excuse me, there are a LOT of articles that need my attention...