Thanks for the review of my Serial Posts plugin. The feedback is useful and you make a good point about user-friendliness of the CSS for the list. I’ve added this to my to do list!

Cheers,

Ade.

Ade’s last blog post..Exclude subcategories from archive pages

Sure, when a single person designs a single piece of software, regardless if it is covers 1 or 3 things, for that one person there’s a much lower probability of miscommunication with himself, and getting the interactions between the all the different pieces of his software wrong, than it is for 3 people who work independently and do not communicate at all. But this is only one side of the coin.

The other is, there’s a limit to the capacity of a single person, be that the knowledge that he may poses , or his ability to master complexity. So even a single person will start introducing an increasing amount of software entropy while growing his software, though usually it’d take more then 3 features to get there.

I think it all comes down to the “domains” covered. Software can do pretty well even with many features when they’re all in the domain of competence of the author.

Alas, competence can only be achieved by doing, so everybody starts as a dilettante at some point. If dilettantes hadn’t be programming, we wouldn’t have a single expert. (I’m telling this to myself whenever I want to blow my mind about some dilettante’s crappy code )

But I’m drifting away a bit into philosophy here.

To come back, I’d make my choice based on whether I feel those 3 things are only variations on the same theme (and hence are in the same “domain”) or not.

Unfortunately, I can’t formulate a general recipe how to decide which is when the case.

In the concrete case of Hackadelic Series though, I felt “collapsibility” would be outside the domain of the series front-end.

Hey, I even managed to come full circle, and back to the original topic

Hackadelic’s last blog post..Sliding Notes 1.4 – Shadowchaser Release

And I don’t mind the topic wandering, it helps raise the tone of the writing here.

In my experience, software that does one thing right is better than software that does 3 things right, mostly because the latter’s “rights” together are usually less, or less stable then each of the formers individual “rights”. This is mostly because when software is trying to do N things right, it usually introduces much more complexity then when N times doing a single thing right. (That is, complexity is usually not linearly proportional to the number and “size” of the implemented features.) At the same time, complexity creates “software entropy”, which again decreases the stability of the software.

I haven’t attended any software project (and I have attended many), where these things weren’t true. This is why “separation of concerns” is such an important principle in software design.

But I see that this goes way to deep into software engineering (as a scientific discipline), and I’m afraid with it I’ve left the scope of your blog, or this article. Bear with me for it – I’m a software engineer by vocation, not only by profession.

Hackadelic’s last blog post..Sliding Notes 1.4 – Shadowchaser Release

In an ideal world, we can have a high degree of code reuse and ease of installation and keep our plugins simple.

I keep looking for these. That last part of the sentence it the problem. Every time someone adds a feature to a plugin, they increase the chances of conflicts and complexity of installation. Of course, the users like lots of features and options. At least I do.

Which is better? Three different plugins, each doing one thing that is related to the other two, or one plugin, doing all three things at once. This is not hypothetical, I know of at least one plugin that fits the description. It is a trade off that developers and users have to deal with.

Of course, without having to deal with this kind of trade off, I would have much less to write about.

And especially someone who values simplicity should be grateful about it. (I’m not talking primarily about myself here ) I consider myself a software engineering expert, and as such, I know that more code usually means more bugs, too. So I always ponder complexity against added value. If a piece of code is used only once – at installation time – but is always present – to an unknown effect – I reconsider if I can get away without it.

As for the “spoilers” – I would have never thought to search for this term. But I did now, and your’re right that most of them are not updated for ages. However I do have a different view on their age: I’d say they are abandoned. As WordPress evolves, old plugins may stop to work (as many have painfully experienced). So, without bothering to check, I would make the opposite bet: There’s a good chance they will soon stop to function.

But don’t get me wrong. I’m good with your view of things. And I don’t try to make you change your vote.

In an ideal world, we can have a high degree of code reuse and ease of installation and keep our plugins simple.

But to get there, we’d need a plugin management system that goes beyond installing and updating individual plugins, but is aware of plugin dependencies, too. Such a system would install a plugin and all dependencies automatically.

And it’s not that such things would not exist already. All Linuxes come with a software package manager that does exactly that.

With WordPress, however, such system has yet to be provided.

Tom, I’ve enjoyed our little discussion a lot, as well as your other writing on your blog. Keep it up.

Hackadelic’s last blog post..Sliding Notes 1.4 – Shadowchaser Release

It is in fact one of the most important things I look at. How easy is it to install? Is it plug and play? Does it just use a widget or is there an option page? Are the options easy to figure out and does the author provide documentation? Anything that adds to the installation, counts against the plugin. It might be a minor point against the plugin, but I will still mention it. And frankly, if something like having to install a second plugin is the worst I can say about a plugin, it is doing pretty well.

And I am sorry to tell you this, but there are multiple plugins with the basic functionality of Sliding Notes. Search the plugin directory for spoilers, both as a term and a tag. The spoiler tags that show up are mostly old, but their simplicity means there is a good chance they will function. And they do not have exactly the same functions as your plugin, but they do hide and show text in a post.

I certainly understand your concern with having to download and install another plugin. But that’s a one-time effort is more then compensated by the increased quality due to avoided redundancy. Afterwards, thanks to WP2.7′s automatic plugin upgrade feature, no significant efforts remain. And with Plugin Central, even first-time plugin installation is a no-event.

BTW, there is a excellent and famous plugin, Similar Posts, whose author has made a cleaver decision to pack code reused by all his plugins into a separate library plugin, so neither it is a completely unusual practice to reuse code that way.

Further, if someone was already using a plugin with a functionality similar to Sliding Notes (which is highly hypothetical, because there is none), there is no need to uninstall it, of course. It is perfectly clear that, to him, there’s no difference if the drop-down functionality was built into Hackadelic Series, or comes from somewhere else. He ends up having the same function active at the same time one way or the other. (To test the truthfulness of that statement, imagine that the Hackadelic Series archive would include Sliding Notes. You still end up with the same code, right?)

And the conflict with Shadowbox JS is not a valid argument here, either, because it is a CSS conflict, and the same CSS would have been part of Hackadelic Series, if it hadn’t used Sliding Notes. The conflict would have occurred nonetheless.

However, I can fully understand the suspiciousness about having too many plugins around. I tend to be suspicious myself about plugins that I don’t know. But suspiciousness is just not indicated in this case.

It was a great discussion though.

Hackadelic’s last blog post..Sliding Notes 1.4 – Shadowchaser Release

I understand your point of view about the utility of using your Sliding Notes plugin for the functionality you are going for. You make good points about redundancy and so forth. But I am looking at it from the point of view of a user that has to install it.

First, it is another plugin to download and install. That is more work for a plugin they do not even want, to make the one the do want work. Second, what if someone has another plugin already installed that has the same function as Sliding Notes? Do they uninstall it and put in yours? More work, especially if they have to go through many posts to remove or change shortcodes. (I have found that having 2 plugins with the same function active at the same time is just asking for trouble. Look at the conflict between Sliding Notes and Shaddow Box, not even trying to do the same thing.)

I certainly applaud your efforts at keeping overhead down on blogs. Plugin bloat can certainly slow things down. But I am trying to look at the plugins for both function and ease of use. I have seen some that are not worth the effort it takes to install them. Yours is definitely not in that area. But from that the ease of use point of view, the extra installation is not a plus.