CodeBard

Making WordPress easy

[CodeBard @codebardcom]

My experience has been that while there may be a few outdated things in the average WPBeginner article, the tutorial itself works. This is no different than the rest of the content google shows. As a result I don't think it's a differentiator. And let's face it - WPBeginner isn't the only site experiencing this kind of traffic drop. So it can't be rankings.

[CodeBard @codebardcom]

Except the 'landing on outdated tutorial' doesnt happen - Google has been deprioritizing non-evergreen content since a long while now, and an article about a software version from 2012 has no way of appearing in search unless the user specifically searches for it. As a result, there is no reason for google to penalize them. It has been able to just ignore low quality backlinks since a long while ago, so it's certain that it can just ignore non-evergreen content that is way too old.

[CodeBard @codebardcom]

@BillErickson Just vanilla WordPress + a caching plugin like WP Super cache makes the site basically static and accomplishes the same thing. And without losing any other functionality.

[CodeBard @codebardcom]

Most membership sites use one strategy and leave money on the table.

Here are 7 monetization strategies that work together to avoid that:

- Early Access
- Archive Vault
- Seasonal Pass
- Tiered Articles
- Premium Login
- Category Pricing
- Content Drip

Each one engages a different type of fan. Here's how to set them all up:

https://t.co/XtKENlf6lp

#contentlocking #contentmonetization #creatoreconomy #creatormonetization #memberpluginpro #membershipplugin #membershiprevenue #membershipsitemonetization #membershipstrategies #monetizationstrategies #paywallstrategy #wordpressmembershipplugin #wordpressmembershipsite via @codebardcom

[CodeBard @codebardcom]

I just gave an example over what would happen if the WP org plugin repo allowed 3rd party Google Analytics. The same principle applies to Woo marketplace or any place where you allow 3rd parties to add scripts. There's no issue with having 3rd parties have listings in any directory. But allowing them to run 3rd party trackers is a legal mess.

[CodeBard @codebardcom]

It's not a matter of plugins - it's a matter of law/regulation. You can't control what a 3rd party does regardless of whatever ToS they agree to. One plugin dev agreed to the terms, even signed a data processor agreement but then turned on 3rd party data sharing in Google Analytics and it got out? The regulators will come after the first party (https://t.co/sbKZsGCKYU), and if https://t.co/ksnC7L1EEX says "But this person agreed to our terms and promised he would not do X", then the regulators will tell you "Ok, you sue them while we sue you. It's your problem"...

[CodeBard @codebardcom]

If any of you were having issues with your licenses & updating Member Plugin Pro, you can try again now.

[CodeBard @codebardcom]

Member Plugin Pro 2.0.1 Brings Customizable Unlock Buttons, Member Display Name Change https://t.co/qA9N8SFowi #creators #memberpluginpro #membership #membershipsite #newversion #news #patreon #patronpluginpro #update #version #wordpress via @codebardcom

[CodeBard @codebardcom]

@jeffr0 The terms and conditions one is something legally inescapable unfortunately - both on the plugin developer side and the user side...

[CodeBard @codebardcom]

Patron Plugin Pro is now Member Plugin Pro and it allows you to run on-site memberships alongside Patreon! Same content, same features, more ways for fans to support you. And setup is simple.👇

https://t.co/68PKhyURdf

#memberpluginpro #patronpluginpro #patronpro #membership #creators #wordpress #patreon

[CodeBard @codebardcom]

I think WPML is the general all around choice for multilingual features still. It used to have performance problems if you publish too many posts in many different languages when you combine it with many tags and taxonomies (results in many joins). But I don't know whether they have already addressed that.

[CodeBard @codebardcom]

Not much difference. Though these keys can use up significant API quota, therefore money. Although, encryption wouldn't make a lot of difference - if the site is compromised, the attacker could have gained access to the encryption key as well. It could only help in the case of a database-dump only attack, if the encryption key was in server env variables etc and the attacker wasn't able to access it.

The AI API providers need to introduce spending caps like some cloud providers to address such things.

[CodeBard @codebardcom]

@johndsaunders @webflow Or just use WordPress. It was built specifically to accommodate non-technical users in using their sites. It already has APIs for anything AI related.

[CodeBard @codebardcom]

What's more, most of those posts are from people who are developers. Not any average user. Naturally, since there is no way a flower shop owner who uses his/her WordPress site to run his business would spend his time managing a 'modern' frontend stack instead of concentrating on his business.

[CodeBard @codebardcom]

@rayhanarif07 @willya_rnd Yep. You can get it out of the box without any effort:

https://t.co/eeXYqfONAN

[CodeBard @codebardcom]

@willya_rnd @rayhanarif07 Here is an ordinary WordPress theme getting 100 page speed score out of the box with absolutely no effort:

https://t.co/eeXYqfONAN

[CodeBard @codebardcom]

People in WordPress are afraid of change because their users - individuals and small businesses - don't like, and don't want change. They use WordPress to run their businesses and sites. Not the other way around. Therefore they have little tolerance for anything that disrupts their lives by disrupting their sites, and with good reason...

[CodeBard @codebardcom]

Well, the biggest problem is still maintaining the code. AI context window easily reaches 1m size even with a small to medium codebase. This can easily be a small plugin. Then the context rot starts, and it's really not something that a non-programmer can deal with. One would need a programmer for that. And in between having an in-house programmer build & maintain such a feature and just using an already well made plugin, the latter would win any time.

The context can be much bigger than code, mind that - if you are coding something that involves externalities (a remote API, services etc), the context grows much bigger than the code itself.

And there is the problem of economics. Even if you successfully manage a middling size codebase with a LLM/agent, you will be burning tokens. A smallish plugin with noticeable context size will easily use a 200k-300k context window and will burn millions of tokens with every small change. With the Western models, this is a cost that no small business or individual can handle. Actually even large orgs are trying to lower their token costs these days.

I don't think context size and economics issues are going anywhere anytime soon. Even if they did, context rot still would be a deal breaker.