On 10/7/07, Aaron Schulz jschulz_4587@msn.com wrote:
I'm mainly talking about higher profile pages. I looked up George Washington. It has around 50 templates and 17 images. That would require MySQL to scan up to 134 rows on each page view to see if things where changed
Not on each pageview. Most hits are on the caches, which are already purged if templates are changed. And you have to load these rows anyway on an uncached hit, since you fetch the template data. So I don't see what the big deal is with also fetching the flagging data.
I'm not convinced at all it's a performance issue. I'll have a chat with Brion about it.
The UI change was not much better, and it even through of Phillip resulting in some confusion earlier.
Yep, in a template vandalism situation. Which is going to be the _exception_, not the rule. The rule we want to strive for _is_ that a page and all its components have been "sighted and current". And giving a clear visual indication of that state is the best way to do that.
But anyway, I'm not arguing we should accept it working inaccurately with templates. I'm arguing it's absolutely worth doing right though.
- What will most people think "revision" and "change" means?
Sorry, but if you're at that level, you've already lost most readers. Seriously. People shouldn't have to care about revisions or changes at all, and most people don't. The point here is to convey in a reader-friendly way that a basic check has taken place. So that I can print the article about a town into a brochure and not worry that it says somewhere in the middle that the residents are all the product of incestuous relationships.
Therefore I think your discussion about the semantics of changes is largely beside the point. From a reader point of view, it's already far too low level.