-
Website
http://www.yongfook.com -
Original page
http://www.yongfook.com/post/view/69/10-reasons-why-i-ditched-wordpress -
Subscribe
All Comments -
Community
-
Top Commenters
-
Andy Heather
47 comments · 7 points
-
katski
22 comments · 1 points
-
Jon Siegel
19 comments · 1 points
-
Wilhelm Murdoch
61 comments · 1 points
-
solace
36 comments · 2 points
-
-
Popular Threads
And, by the way, I totally agree with you but not everyone masters PHP to a level where they can just create their own blogging platform. THAT is the beauty of WordPress. It's just so bloody accessible to the non-geek peasant.
I really want to know how you fetch stuff from Tumblr btw.
I'm using the API to pull in data items into my own database and then displaying them with a custom built front end app. If you just look at the source of my site you'll see it's not just a simple embed.
I'm sure the way you did it makes much more sense. I just try to find solutions around my own limitations.
Looking forward to seeing how you pulled it off :)
All will be explained soon. The solution is still quite simple, but it's definitely not as simple as just adding an embedded script, unfortunately!
maybe this weekend I'll put together a tutorial.
I think it really depends on how serious you are about blogging and what your level of technical expertise is. I'd generally recommend a hosted platform as they're getting better and better these days and they are even customisable/plugin-able to a certain extent - but then again I'm biased (disclosure: I'm lead technical architect on Webjam which is an (MVC based ;)) hosted cms/blogging/social networking platform). If you're more technically minded, then wordpress or even a custom system like you've written can probably suit you individual needs slightly better, but this certainly isn't the case for most people.
Marcus
http://www.webjam.com
http://www.chickerino.com
Thanks for your comment. I don't want to get into some silly pissing war here, but I do feel you are making an unfair comparison. My individual blog, compared to a hosted platform that has to serve hundreds (thousands?) of different blogs each with their own audiences and traffic levels will of course have to have a massively different scalability strategy.
Anyway, I agree - caching is not the be all and end all, but it's a decent enough solution for an individual's blog like mine.
Also, when trying to make a page load more efficiently one of the areas you can look at - and it's disingenuous to deny this - is making queries snappier. This involves a lot of trial and error - for example, can I make one expensive query in the place of 5 separate queries, and is that expensive query going to hit the database more or less than doing it all separately.
My point about a page with many, many queries on it was more about the carefree attitude to plugins that Wordpress kind of encourages you to foster. Your blog could begin as a model of efficiency but as soon as you start adding all these plugins written by unknown people you open yourself up to serving a page that has many more db hits than you intended, and sometimes these can be very expensive requests (a popular addition is one of those "related posts" plugins - these in particular are really fat queries) which just add up and up to the undisciplined or uneducated blog owner and make your pages very inefficient.
I think part of the reason that Wordpress plugins are so inefficient is like you say that they use so much code and are not part of an MVC framework - which would certainly more easily facilitate the re-use of certain bits of data without having to re-query the database. The other problem that Wordpress has is that it's simply old. Therefore it has the legacy of having to support (or not break) all of the old plugins when they upgrade. This really hinders their ability to do a proper 'modern' re-write of the whole system.
I used ExpressionEngine on a client website and noticed issues similar to your complaints about WordPress; particularly the number of queries performed on each page load. Thankfully though, the Ellis Lab folks have done a great job with that product, allowing you to use various caching methods and turning off certain db queries at a pretty granular level.
Your posts on this and the lifestream have me more inspired about design and dev possibilities than I've been in 2+ years. Thank you for the lovely nuggets of inspiration--I am beside myself with glee :)
I recently tried to upgrade to 2.3 and it borked on me."
Ah.
Your homemade theme, huh? Too modified to debug and fix? The "Windows" of themes?
"Wordpress is quite the resource hog."
There may be some other problem you haven't diagnosed. Looking at Compete/Quantcast/Alexa, you have about 2 percent of my traffic, and I'm not having any load problems.
Lets brush aside the fact that Alexa/Compete/whatever 3rd party
analytics tool that requires all users to have a browser plugin to
calculate numbers are horribly skewed - because I'm not going to sit
here and pretend my blog gets mad traffic. I have, however, worked on
a number of very high traffic sites where we either rolled out
production sites based on wordpress or did some A/B tests and found
that Wordpress dies very easily without fat resources behind it. You
can use all the fancy optimisation plugins you want, the fact of the
matter is Wordpress as a "framework" is quite heavy. So with this
knowledge, I don't really want to use Wordpress to serve the front end
of *anything* I do from now on - not because I expect huge traffic on
all my projects, but because if I'm able to put together a more
efficient, faster-loading front-end alternative*, why the hell
shouldn't I? Not a particularly hard decision there.
As for the other point, the uprgrade bork had nothing to do with my
theme. It just totally destroyed my schema (I think at some point
from 2.0 ~ 2.3 they completely changed the schema, to introduce the
new "terms" concept) and I lost all my categories, they just got
erased in the DB. Yes, I had backups, yes I tried it again and again
- but really once you've wasted any amount of time on that kind of
thing - just dead time because the software didn't do what it was
supposed to do - it's very frustrating. And from that point I started
to think about rolling my own software that I would have more control
over.
*I still totally see the value in using WP as a CMS though, and if you read the article all the way through you'd know that. I just think it does a pretty meh job at actually displaying content. You can write a simple MVC app that pulls data for display from the WP db very easily, and will be more efficient and easier to maintain (i.e. you write your own extra functionality using standard class files in a standardised application structure instead of getting into a hell of using all these random plugins shoved in your plugins folder) than a WP theme. Of course, this isn't a solution for everyone - I'm not suggesting everyone drops WP this instant. But definitely people who have confidence in their programming skills can adopt this approach, which as I said, is how I tackle most client work these days. WP backend with a custom-built, highly maintainable and extensible front end application built in a more efficient framework.
Great blog entry by the way YF! Keep going!