To fork or not to fork - that is the question

Posted: Sun Mar 14, 2004 2:16 am

I guess I phrased that wrong, or badly... or something.

I know that you can buy nuke from phpnuke.org... what I was referring to is a "mature" (as you phrased it so accurately) version that is at least stable, and predictable. I don't think that his code is at all mature, regardless of what version it is. I start with that code, but I patch it considerably. The very people he is yelling about are actually the people that I prefer to get my code from. No offense, but I need stable, and that ISN'T what I get from phpnuke.org.

Mambo and e107 are close, but not close enough. Again, they are not exactly what I would call "feature rich". Bad? Certainly not.

Thanks to NukeCops, Protector.. and well.... ME, I find my code is pretty stable in that they have fixed the major holes, and I have patched the little ones.

While I know that you haven't officially "Joined Forced" with anyone, I'm neither bothered nor joyful about that fact. I believe that cooperative efforts are great, but I also believe people are people and sometimes "joining forces" isn't a good thing. Again, I like what I have seen here and am pretty darn happy about some of the scripts you have shared (see my posts!) so I'm not complaining about anyone's situation, especially yours! I am saying that if I had you, chatserv, and a few others working from my shop, I know we could put out a d*** fine version of nuke that people would be more than happy to shell out $100.00 for! I'm also sure I could sign them up for $200.00 per year support contacts. I mean, we're talking $200.00 LESS (in total!) than what MS charges for a 5 user version of portal for 1 year! And you can't call what they offer "support".

I think that nukecops and others (I'm trying NOT to mention names here) have made attempts in the past to identify holes in the code (AND CODING METHODOLOGIES) used by Nuke authors. If some have taken this badly (nukelite), well.. it happens!

My comments were really intended to identify that IF there was a "Mature" version that was being evolved by a group of expert Nuke sources, that included EVERYTHING that I've had to patch together from about 20 different places, I think that it would be sellable as it has the value of being something that a webmaster can't find or have supported via a single source.

I think the community has asked for THIS, but has yet to find it.

I think that NukeCops/NukeFixes/NukeResources have come close to this, but lack some of the finishing touches in that they are (in my humble and PERSONAL opinion....so don't go off on a rant for me saying this!) supporting too MANY old versions of PHP-Nuke. I can understand HOW, as it is real hard to upgrade with new "revisions" coming out every month, some without a clear upgrade path and many with "historical bugs"... meaning, you can't upgrade without re-creating bugs you've fixed through every OTHER upgrade you ever performed.

I think when you are forced to support multiple versions (that is to say for example the people at nukecops, nukeresources, nukefixes) that have many differences (the various versions of nuke they are supporting), you end up missing things in your own revisions.

I (again, personal opinion) think that publishing accurate code is more important than simply putting out the update. It bothers me when people put fixes out that haven't been tested for the platform THAT THEY ARE RELEASING IT FOR, knowing (and knowing is the key word) that there will probably be issues due to the differences between what they RUN and what they are patching.

It's the old adage, you can't really SEE if you've fixed it, if you aren't running it.

The other downside is, who the hell wants to run 20 different development servers with 20 different configurations of patches to support 20 different versions of Nuke (IE.. 6.4, 6.4 post-fix, 6.5, 6.5 post-fix, 6.6, 6.6 post-fix, etc..)? Certainly not me!

Then there are the script writers.... I won't be bashful about my thoughts here either. The ones who give us mods, and cool new features ... TOYS! (ravenphpscripts.com!) on top of some fixes, without some of these scripts, I would NOT be a happy camper. Let me take this opportunity to talk about "INCOMPLETE ENROLLMENTS!" I LOVE THIS SCRIPT! Laughing

http://www.ravenphpscripts.com/download-file-59.html

I'm going to keep raving about this script too!

Anyway.. I hope that clears up what I AM (not was, but AM) trying to say.

I think there is a market. I think people WANT to pay for a "mature" feature rich product that includes support and custom graphics design. Is it out there? If so, I haven't seen it yet. Again, some have come close, but I don't think anyone is "nailing" it. Let me know if you start selling a turnkey commercial version, I'd love to look at it.

But that's just my humble opinion.... That's OPINION (for those that are about to yell at me!)

Oh.. btw... I want to take a second to thank chatserv for my SQL command to automatically inject the code for my new "firefox" statistics modification. Another example of getting a little help, which is a good thing... something that I can't even think about getting from phpnuke.org, since there aren't any forums!

Site Admin/Owner Joined: Aug 27, 2002 Posts: 17088

Steph,

I seem to have such a natural way of miscommunicating anymore Confused

. You phrased NOTHING wrong nor did I misunderstand anything. I think I'm just very much more empassioned about this right now and my writing is reflecting that. My statement about not joining forces was just meant to say that Chat nor anyone else had 'gone over to the dark side' by joining this avent garde approach Smile

and to keep anyone from miscommunicating their 'allegiance'. Anyway, don't want to labor this. Your points are all valid, opionated, articalate, and welcomed!

Posted: Sun Mar 14, 2004 8:19 am

paranor wrote:

Since I can't do anything important I'm willilng to purchase a couple of books or anything if that helps. Just let me know.

Nothing important? Keep those new accounts coming in Smile

- I find that etremely important :lol:Thanks again!

Posted: Sun Mar 14, 2004 8:34 am

paranor wrote:

I was going to make a joke that the MCP is evil and that you need Flynn to get rid of it. Smile

Ah yes - another junkie from another age Laughing

Posted: Sun Mar 14, 2004 8:46 am

Rikk03 wrote:

Hey Raven, Could you PM me the new name - give me an idea for an domain name url for a support site?

Richard

http://php-openportal.com (Original Laughing

)- Nothing is there - yet Smile

except referral links.

Posted: Sun Mar 14, 2004 2:12 pm

Raven,

I have spent this afternoon and this evening reading through the post that you cannily saved from phpnuke.org and educating myself in the why's and wherefore's.

It would appear to me that a small minority of people are concerned with search engine friendly URL's. The majority just want a working CMS. This is the same whichever CMS you use - postnuke has exactly the same issue (friendly URL's).

You are talking about a performance trade off for producing HTML output - but I would much rather a CMS that has high traffic but you need to spend a wedge extra on faster hardware. Than a CMS that performs admirably but has little traffic?

With that in mind.

Am I right in thinking that with a bit of work (all Laughing

) I need to do is understand the way that mod_rewrite works, use a tool like this http://www.webmaster-toolkit.com/mod_rewrite-rewriterule-generator.shtml and I can design a search engine URL rewriter for any PHP CMS?

My point being that who cares about performance issues because by the time your site is pulling enough traffic to worry about it you can afford to pay a decent programmer to optimise the code or spend the money on faster hardware.

So if I use mod_rewrite to produce my friendly URL's then I wouldn't have to worry about which CMS I use - is that correct?

The nexgen project at audioslaved appears to have died a death, however if I just take the existing urlin and urlout parameters and place them in .htaccess then this should work a treat? Without worrying about how it was optimised? (the nexgen way).

This would mean that if I learn how to use mod_rewrite then moving to a new CMS would hold no fear for me.

Is this logical?

Thanks
Tony

Worker Joined: Aug 28, 2003 Posts: 227

Raven wrote:

paranor wrote:

I was going to make a joke that the MCP is evil and that you need Flynn to get rid of it. Smile

Ah yes - another junkie from another age Laughing

Whew. I was beginning to wonder if I was the only person old enough that knows about that movie. Smile

Posted: Sun Mar 14, 2004 2:38 pm

Storebuilder, You make it sound so easy!

I know you are talking to Raven, but I just want to add my 2 cents.

Quote:

It would appear to me that a small minority of people are concerned with search engine friendly URL's.

I don't really agree with that. I think in the future (if not already) having this feature will be a "KEY" element. If for no other reason, then to prevents bots from "hanging" or "slurping bandwidth" when they get to a site with .sid's. If you miss them (by not watching your Awstats for example) then they can suck enough bandwidth in a week for many people to reach their limits for the month.. and that's without having ANY actual USERS hitting thier pages! Again, I think the solution will be to have the engine be PHP and the output be in .html templated pages.

Quote:

You are talking about a performance trade off for producing HTML output - but I would much rather a CMS that has high traffic but you need to spend a wedge extra on faster hardware.

My beta box has dual 3GHz CPUs with 2GB of RAM and 15K-RPM SCSI drives, how fast do you think a server needs to be? It jumped from 15% utilization to 88% with GT enabled with only 1000 pages of index data.

Quote:

My point being that who cares about performance issues because by the time your site is pulling enough traffic to worry about it you can afford to pay a decent programmer to optimise the code or spend the money on faster hardware.

It's still a cart and horse issue. Even if you are getting indexed well, the LAG kills you. People don't like slow loading sites and GT makes it lag! The more users online, the worse it is.. which is not good. Also, you are forgetting the GT locks you in.. once you've enabled it, you can't change structure without creating a huge amount of 404 problems in the search engines. If your locations change then you are going to get slammed by people going to bad addresses.

Quote:

Am I right in thinking that with a bit of work (all ) I need to do is understand the way that mod_rewrite works, use a tool like this http://www.webmaster-toolkit.com/mod_rewrite-rewriterule-generator.shtml and I can design a search engine URL rewriter for any PHP CMS?

In fact GT will work with almost ANY php based site. While again, I think you make the process sound overly simple (which it isn't) I don't think that mod_rewrite is the real solution at all... Even Apache doesn't think mod_rewrite is the solution. That should say something!

Quote:

So if I use mod_rewrite to produce my friendly URL's then I wouldn't have to worry about which CMS I use - is that correct?

You will still have to worry about WHICH version of GT that you use and also you will have to worry that if your Content Management System doesn't have all of the features that you want (in terms of modules and/or Blocks) and you EVER change, you will get a zillion (I love that word) 404 errors. You'll end up having NOBODY coming to your site because every time they see your domain name, they will assume the link is bad!

The only solution would be to re-open under a totally new domain name!

As for the rest of your post, I would worry. I would avoid implementing GT if you can in any way do so. Remember, once it is done.. IT IS DONE, going back is almost (if not completely) impossible. If you are not going to enable GT features for truly dynamic content (like your phpBB for example) then stick to META. It may take longer to propagate, but it won't create the nightmarish situation that enabling and being stuck with GT can. Also, remember that if your structure EVER changes, you're toast. So what do you do when the next security update comes down the road that in any way (shape or form) changes ANY of those pre-defined and indexed destinations? Or what happens if your phpBB database gets corrupted and you have to re-index and the message numbers change? I would worry. Again, I think GT is a great idea for starters, but implementation has identified flaws using it. A new solution must now be contemplated, designed and implemented to achieve the output results, but instead of by conversion, by internal design of the CMS.

Just my 2 cents, Pass the salt!

Posted: Sun Mar 14, 2004 2:46 pm

WOW ... that FB quote is really, really sad. "Go away and do it yourself" attitude from a developer, isn't at all helpful for anyone. The ego thing (hurt?) is sooo obvious.

Raven ... your plans for new OO system are dancing around in my head. I've read all the posts in this thread and am still thinking about a contribution I could make. Either in html/gui stuff, or graphics, or support. Two days ago I returned from a 15,000 kms, 5 week road trip, and I'll be settling into my happy nuking environment any second now.

Coincidently, I'm installing E107 as on a test site today. Six weeks ago one of my friends thumbed her nose at my nuke7 offering and she nuked the nuke! Then, in a few hours there was E107. I taught her everything she knows... so I'm checking this out for sure.

If I can make my webportals do what I want them to (have features and security), without all those file changes mucking up my previous changes, then I'm going to be really, really happy. I've got Calendar, Protector, etc working on my main site, I've modified my mods a hundred different ways over the past couple of years. Just this morning I attempted to install your Stories Order mod, again. This time it worked for me! Smile

That mucking around in the files ... I do it because it has to be done, like Chatserv, if I'm working on php I'm happy.

But I'm not a trained php (or any other language) programmer, so getting the support when I run into a problem ... hasn't always been easy. And the one upgrade I did, was like a nightmare. I've still got some bugs and the solutions are not easily found. Easy plug-ins capable of being finessed and massaged by the SuperNuker sounds like nirvana. A friendly support site, or web of sites, is going to be essential when Nuke-OP is the portal of choice.

I think FB is excusing his damaged ego with his percieved behaviour of Nukers. Developers/Sysops need to have ducky feathers so all the water (users) can run off their ducky backs, without personality breakdown. It's sad. Because the questioning, demanding, and silly Nukers can create excitement/mods from the Developer's vision.

Posted: Sun Mar 14, 2004 3:22 pm

storebuilder wrote:

Am I right in thinking that with a bit of work (all ) I need to do is understand the way that mod_rewrite works, use a tool like this http://www.webmaster-toolkit.com/mod_rewrite-rewriterule-generator.shtml and I can design a search engine URL rewriter for any PHP CMS?

So if I use mod_rewrite to produce my friendly URL's then I wouldn't have to worry about which CMS I use - is that correct?

Is this logical?

Thanks
Tony

Overall, I would agree. However, I lost a very good account because I was hosting him on a shared server. It had the bandwidth but his site became very popular, not in terms of quantity of members necessarily, but in content and members postings. Apache was creating so many threads that it locked out everyone else and then the other accounts and traffic started queing up and eventually the cpu usage shot to 88%! I hated to lose him, which by the way is the only account that RavenWebHosting has ever lost Smile

(shameless plug) but my host threatened to kill his account if I didn't. And yes, it was GT - I proved it. He was not able to hang on another month as I was in the process of getting my dedicated Dual Xeon w/HT and we both understood that it was the only way. My point (yes there is one Smile

) is that he wasn't making enough money to cover what he needed, per se. I am in the hosting business to make a little money, but more to offer a reliable hosting service with real hardware AND software support, so I upped the ante and put enough horses in to pull that GT cart Wink

- If I build it, they will come!

But yes, regardless of which CMS (actually regardless of the application) you use, mod_reqrite offers a plethors of possibilities and opportunities! But like any powerful weapon, misuse can hurt you.

Posted: Sun Mar 14, 2004 3:57 pm

Ok - what I was thinking, and you obviously know this being in the hosting business, is that actually very few sites get worked on daily and quickly become a runaway success.

So my train of thought ran that if in six months time I was going to change CMS - because something better had come along, then I would want a quick and dirty way of getting a site indexed - until a better solution was found or integrated.

However, for the majority of users mod_rewrite would be an adequate solution - because they will never draw significant enough traffic to worry about it.

So you see, it now does not matter to me that search engine optimisation is not high on your list of priorites. I, and others, can get by until a better solution is found.

So put me down for the team.

Posted: Sun Mar 14, 2004 4:35 pm

64bitguy wrote:

I think that NukeCops/NukeFixes/NukeResources have come close to this, but lack some of the finishing touches in that they are (in my humble and PERSONAL opinion....so don't go off on a rant for me saying this!) supporting too MANY old versions of PHP-Nuke.

I can't speak for everyone but i can comment on what relates to me and the last 2 sites mentioned, i could support only the latest version but we have to be realistic, not everyone keeps up to date for one reason or the other. If someone came to me asking for help on PHP-Nuke 6.5 and i told them i can't help if they don't upgrade to 7.1 we'd have a lot of dissapointed people in the community, i still support 6.0 because one of my sites was still using it until recently and because i know many others do as well.

I'm one person and as such i do what i can but i'm sure you already know that, same thing applies to many of the people that devote time to helping others, we could dream about having everyone unite for the better of the community with all working in the same direction but we all know that hasn't worked in the past and i honestly believe it won't in the near future, the closest one can get is to put together a team of a few and hope for the best.

Hangin' Around Joined: Feb 20, 2004 Posts: 29

wraith wrote:

First, these topics are missing or have problems in PHP Nuke (for me):
1) No documentation at all (well, a book in german are on its way......cant read german so....).

I was away from this thread the last days (and probably will be for the coming ones) due to a cold I caught - but I *had* to correct you on this: the first book on PHP-Nuke is already there, in english and under the GNU Free Documentation Licence - click on this site on "Nuke Manual" (in the "Site Navigation" box to the left) to see its PHP-Nuke module version:

http://www.ravenphpscripts.ravenwebhosting.com/modules.php?name=PHP-Nuke_HOWTO

Get your preferred format from

http://www.karakas-online.de/EN-Book/formats.html

BTW, Raven, I'm on part 1 and 2 - we'll see for the rest. Please make this GPL.

See you all soon

Chris

Client Joined: Jul 18, 2003 Posts: 977

Raven, I came into all of this quite late it seems. While I cannot "do" anything with coding or designing you know that I am here to help in any way you can see to put me too. I believe you know my level of Nukeness so... Smile

Whatever you need do not hesitate to ask.
Now for a few thoughts, one of the things I have always feared and faced with some dread is updating to a new version. I have always been afraid something would be lost in the transition... so I hope that updating to your system will be easy for the first time and then easier for any subsequent updates. The other thing I have always hated was hacking the code to get something to work, from what I have read it looks like this is all going to be fixed from the ground up so kudos to you! I am excited for this. Whatever you need, do not be hesitant to ask of me. God bless your endeavor!

New Member Joined: Mar 01, 2004 Posts: 8

I believe phpnuke is a big success because they are people like you, chatserv, Humpa(where is he lately?), Chris-au and Chris(Karakas) that have helped, learned and guide people like me editing or debugging our phpnuke website.
You know from my view FB greatest think was that he managed to do THAT not phpnuke as it is!
So i believe that without that help the project will not go any father!
If you believe that you can manage to do the same think i believe you should DO IT.
I dont care if you give the project a new name or not just make it easy to modify and secure

P.S.
There is No Doubt that the PHPNUKE project has nothing to do with the project that it was in the beginning

P.S.2
Sorry for my bad english i am from Greece... Very Happy

New Member Joined: Mar 08, 2004 Posts: 9

Quote:

You know from my view FB greatest think was that he managed to do THAT not phpnuke as it is!

Not wanting to start a flame war here, but FB didn't do that. We did. Yes, I mean we including myself back when I was still a member of the PHP-Nuke community.

All FB does and has ever really done on PHP-Nuke is add code to it. To my knowledge, he's never supported it himself, rarely done his own security fixes (most that i know of have come from Chatserv over at NukeCops or BobMarion from NukeScripts Network plus a few others, but without looking them up on various forums online, I can't tell you who or what they fixed)

Raven's taken on a big project, and I don't envy him that, even if I am on a project other than PHP-Nuke myself. From what I've read, he's going to start a whole new CMS / Portal, not produce another fork of php-nuke (or at least that's what it sounds like if the programming for Raven's ideas is really as extensive as I think it is).

AI

Posted: Mon Mar 15, 2004 4:53 pm

You are correct AI. It is not a fork.

Posted: Tue Mar 16, 2004 7:16 am

Just as an update, I have been really concentrating on investigating the feasibility of whether a nuke 'rewrite' is the way to go. Even though I desired it from the beginning, I needed to give it a thorough investigation. Because of the way nuke started (Thatware) and then has been built, patched, paper-clipped, etc., it would still need so much rework that maintenance, one of the Prime Objectives, I do not feel that is the best way to go. Now, for the $64 question: Will it be backwardly compatible? I doubt it. But, as we design it, we will try to have an upgrade/conversion path of some kind. As I have said previously, I am intending to build a framework to plug things in. In addition, I hope to build it with some pre-defined 'structures'. Structures will be analogous to 'Nuke', 'Std', etc.

I realize that this vision is grandeose, blue-sky, 'the man is out of his mind', etc. I also realize that 'many have tried and many have died'. Quite frankly I want to be among the many that tried Smile

. Keep checking back for updates to the project. It is not stalled, it's just being planned and worked. Thanks for all the feedback and support!

Posted: Tue Mar 16, 2004 2:29 pm

Quote:

Quite frankly I want to be among the many that tried

Quite frankly I want YOU to be among the many that tried Very Happy

Quote:

Keep checking back for updates to the project

That is one thing you surely don't have to ask Very Happy

Posted: Tue Mar 16, 2004 3:15 pm

Speaking for myself - as long as I could port over the essential content - i.e. forum posts and news articles then that's all that would really bother me. Oh - and members too - that would be pretty important Rolling Eyes

I spent so much time taking my reviews from nuke reviews to MReviews to Emporium to reviewpost. Very Happy

I'm used to moving home Very Happy

I think I would speak for most users in wanting the core content ported.

Raven, if you are like most programmers I know - and you are probably not Very Happy

then I know you would want to do this from a blank sheet.

Can you give us a clue from the outset so that everyone knows where they stand as to how much "backward Compatibility" you envisage.

If you are going to start from scratch, then I think everyone here agrees that it's the correct approach.

What is crossing my mind is how are you going to gain supporters quickly, for a project that is not promising backward compatibility?

Do you need a marketing guy? In that instance I may be able to help Very Happy

Posted: Tue Mar 16, 2004 3:21 pm

Doesn't everyone? Wink

Posted: Tue Mar 16, 2004 3:35 pm

Yes on marketing - never travel w/o one Laughing

Storebuilder wrote:

Can you give us a clue from the outset so that everyone knows where they stand as to how much "backward Compatibility" you envisage.
...
What is crossing my mind is how are you going to gain supporters quickly, for a project that is not promising backward compatibility?

Excellent question and I even have an answer Smile

, but it requires one to "think out of the box", and you hit the nail on the head in your previous post.

When you look at migration/portability issues you have to look at basically 2 main issues. Preservation of functionality and preservation of data. In most scenarios, preservation of data would take priority - and it will. In my design, I believe you will have much more functionality than you do now. Does that mean that you will be able to plug in addonX and hack99? Probably not. But, hopefully the framework will make it much easier to accomplish the functionality. And, there will be things built into the core that are addons now and have to be redone every time there's an upgrade!

So Mr. Marketeer, I have put you down for the position and have given you some material already. Now, do your thing, sell it based on a concept, and then we will build it. Hmmmm, isn't that VW (vaporware)?

Posted: Tue Mar 16, 2004 4:00 pm

sixonetonoffun wrote:

Doesn't everyone? Wink

Of course - Can I test this quote thing, because I never fully understood it Smile

New Member Joined: Mar 17, 2004 Posts: 6

Sigh. Another fork? How many are there now? How long until every other person who has worked on PHP-Nuke in the past start their own version? All that scares me about this is their being so many forks that no one knows which one to devote time into creating new modules for. I have been running my Nuke site for a long time time and really do not want to have to port all the data over to a possibl incompatible system.

Posted: Wed Mar 17, 2004 3:17 am

Peejay. maybe u should read the posts Wink

it's not a fork, it's a whole new portal.

AI