Should products that only perform template edits be in "Add-ons"?

[Lea Verou]

Originally Posted by Alfa1

Maybe the terminology is not optimal for the categorization, but the current categorization divides modifications into those with and without manual template edits. I think this is optimal for novice users and those who want to avoid manual edits in relation to the hassle of them when upgrading vb.

That's not correct. There are a lot add-ons that include manual template edits (but do add functionality to the forum, and that's why they are in "Add-ons").

[Alfa1]

Then that doesn't make any sense.

[Lea Verou]

Originally Posted by Alfa1

Then that doesn't make any sense.

What doesn't make any sense?

[Lizard King]

Originally Posted by Wayne Luke

Same... Automatic template edits tend to cause so many problems if you change a style even a little bit. Making manual edits is a lot easier when you have a lot of edits in your templates across the board. From my point of view, they make providing support a lot worse as well, simply because you can't easily get to a default style without disabling functions on a customer's board.

An addition to this automatic template edits use more resource on your server.

[Dean C]

Originally Posted by Lizard King

An addition to this automatic template edits use more resource on your server.

Not if they are done efficiently. Of course they technically use more resources, but it's such a miniscule amount it barely makes a difference even on large sites

[Lizard King]

Originally Posted by Dean C

Not if they are done efficiently. Of course they technically use more resources, but it's such a miniscule amount it barely makes a difference even on large sites

If you have a lot of plugins or products installed on your board it makes difference and imagine if you even donot have enough server resource then disabling these auto template edits may be a saver for you

[Dean C]

Originally Posted by Lizard King

If you have a lot of plugins or products installed on your board it makes difference and imagine if you even donot have enough server resource then disabling these auto template edits may be a saver for you

It makes absolutely no difference how many plugins or products you have installed if you're dynamically changing the HTML before its output. As long as you do it carefully, and efficiently it barely touches the resources.

[TECK]

Originally Posted by Dean C

Not if they are done efficiently. Of course they technically use more resources, but it's such a miniscule amount it barely makes a difference even on large sites

Dean, are you sure? Let's do a quick test: Emulate on your test server 1000 users that do 5 preg_replace() per second in a timed loop (through an eval() function) and let me know what are the results.
Just for your information, using eval() for templates is already bad... now imagine that you stiff your code with the replaces, especially the ones that use regular expressions. Very bad. So for me, this is definitely a no go.

Originally Posted by Dean C

It makes absolutely no difference how many plugins or products you have installed if you're dynamically changing the HTML before its output. As long as you do it carefully, and efficiently it barely touches the resources.

Let's put it this way: most vBulletin developers hate to use the plugins, for one simple reason: performance. The more code you add through plugins, the worst your board will run. This is not about writing judicious code, is about pure and simple math. More is worst on a busy server environment. So ya... adding more plugins will ruin your board performance.

We don't want to get into those details related to performance tests. vBulletin devs have an impressive array of server configurations and they test their software from any possible angle and performance, which I guarantee you... no hacker does it on those forums. I'm a little freaky with server software (to a point that I write my own custom RPM's to improve the overall server performance) and I still don't have 25% of the capacity to test things like vB. I only have 3 servers to test a piece of code. 3 servers that hammer one board emulating thousands of users spinning like crazy your disks. Still this is not satisfactory to some customers. You get the idea? So for sure I will not edit in any way, shape or form the original vBulletin board, unless I can prove it 100% that my code will not break anything.

You talked about "doing it carefully, and efficiently". I'm just curious, what is your testing process. Until you reply, let me tell you my coding method also: I edit as less as possible the actual vBulletin files.. and if I'm forced, I never use the plugins sytem. I aways edit the PHP code, it improves the code performance by a lot, compared with the same code used through plugins. That is tested by myself into heavy server environment, so I know is true.

[Dream]

Originally Posted by TECK

I never use the plugins sytem. I aways edit the PHP code, it improves the code performance by a lot, compared with the same code used through plugins. That is tested by myself into heavy server environment, so I know is true.

That's a good point. But another point is that computing resources should be used to make life more pratical, as in not needing to edit files everytime you have a new version of the software, just upgrade and everything still works, saving you a bunch of time you can spend on other things.

[Lea Verou]

Originally Posted by Dream

That's a good point. But another point is that computing resources should be used to make life more pratical, as in not needing to edit files everytime you have a new version of the software, just upgrade and everything still works, saving you a bunch of time you can spend on other things.

Agreed
And anyway, I would prefer to pay more for a better server than to have to edit the files on each vb upgrade

I think we are quite off-topic...