Page 1 of 1

custom forum or proven software

Posted: Sat Sep 22, 2007 4:31 am
by s.dot
This is a hypothetical situation.

Say you own a site with a custom forum that has over 1.1 million posts and 40,000 topics made. It is finely tuned, the site likes it, and it's working well.

I ask of you the pros AND cons of switching a custom forum to a forum software such as PHPBB or VB?

I have my list of pros and cons, but the question's not really for me. I may decide to show this topic to a client, if the answers are sufficient. (makes the question not so hypothetical, huh?) :P

Posted: Sat Sep 22, 2007 9:30 am
by superdezign
You don't switch from a working solution unless the working solution no longer works. If they are constantly running into glitches and problems, or they have difficulty with moderation or get a lot of complaints, *then* they switch. However, if there are no problems, there's really no reason to waste their time on the switch.

Posted: Sat Sep 22, 2007 10:40 am
by feyd
Yeah, I'm liable to go with the American saying "the squeaky wheel gets the grease" here .. or alternately, "if it ain't broke, don't fix it."

The only thing I'd consider is the security hole situation. Granted, with entirely custom software an attacker may, potentially, need custom software to attack it.. but if it's built in a generic enough fashion, all the existing tools will work just fine.

Posted: Sat Sep 22, 2007 4:27 pm
by s.dot
Yeah, but *hypothetically* the switch is going to happen.

What would be the good points of it
What would be the bad points of it

The client is 50/50 on doing the switch.

Posted: Sat Sep 22, 2007 5:03 pm
by feyd
What points do you have so far?

Stuff that comes to mind quickly: Existing software
  • Pro: Tested on more varied servers and configurations, often leading to more robust code. Thereby often more secure.
  • Con: Often a slower development curve, but more stable.
  • Pro: often easier to modify and integrate modifications, or at least probable to have more modifications available that don't need too much rigging to get integrated.
  • Con: quite generic in its default installation and often takes time to tweak "just right."
Overall, custom would be the opposite on many of those points.

Posted: Sat Sep 22, 2007 6:56 pm
by s.dot
My list consists of all pros for keeping the custom software.

- "If it ain't broke, don't fix it"
- Search Engine Optimization could be done much better than the millions of clone forums on the internet
- Much easier to customize, rather than searching for a specific mod or pre-existing code
- Huge hassle importing 1.1 million posts (with custom made bb, etc)
- Huge hassle integrating into the current system
- Identity. Lots of companies I presume pay big bucks for custom made software.

This is just my list. I really don't see any pros to switching for no real reason other than to switch. I told the client this but wanted to have some backup, or some other views presented other than my own, so he can make a good decision.

Posted: Sun Sep 23, 2007 12:30 am
by mrkite
Let me put it this way:

there isn't a single opensource forum software package out there that hasn't had severe vulnerabilities.

Posted: Sun Sep 23, 2007 6:51 am
by The Phoenix
scottayy wrote:My list consists of all pros for keeping the custom software.

- "If it ain't broke, don't fix it"
- Search Engine Optimization could be done much better than the millions of clone forums on the internet
- Much easier to customize, rather than searching for a specific mod or pre-existing code
- Huge hassle importing 1.1 million posts (with custom made bb, etc)
- Huge hassle integrating into the current system
- Identity. Lots of companies I presume pay big bucks for custom made software.

This is just my list. I really don't see any pros to switching for no real reason other than to switch. I told the client this but wanted to have some backup, or some other views presented other than my own, so he can make a good decision.
I'll play devils advocate then.

- It may be broke and you don't realize it. Slower, bad queries, incorrect (x)html, bad integration, etc
- Many existing forum software packages have had extensive work done on improving SEO, including pre-made mods, and papers on how best to do it. phpbb comes to mind in a big way here.
- Searching for completed code is easier than coding your own? That doesn't sound accurate.
- Hassle to import the posts, but perhaps afterwards, it will be easier to work with them, since most opensource forums have substantial work completed on dealing with your posts. Whether its export filters, mods, etc.
- Integration may be easier in some areas, and harder in others.
- Lots of companies pay big bucks to gain favor with the open source community, where running the software and offering up a few patches back to the main developers can do the same for free
- You can find a phpbb developer, or an ipb developer, relatively easily. Finding someone that has experience with your custom coded software? Looks like its just the people you already know.
- Code quality. Having external folks look at your contributions will usually improve the quality, since not everyone tests on 16 databases, 12 web servers, and seven operating systems.
- Learn a marketable skill. If I know you've worked on code I am familiar with, that means something to me. If you've worked on custom code, I know very little about your skill.

I'm actually in the 'if it ain't broke, don't fix it' camp, but since you asked for the opposite viewpoint, there are some arguments for it.

Posted: Sun Sep 23, 2007 6:54 am
by The Phoenix
mrkite wrote:Let me put it this way:

there isn't a single opensource forum software package out there that hasn't had severe vulnerabilities.
And odds are, so has the closed source package, but with no awareness. As a result, you didn't have dozens of developers contributing a fix that came out in hours or days that solved the problem (the right way) on a huge variety of platforms.