Joomla! Discussion Forums

The best suggestion I can give for the SEO section of the forum would be to post a request in the Sites & Infrastructure - Feedback/Information forum. And, I am thinking *right now* given all the irons in the fire, might not be a good time. But, I have no idea - make your request known there. The other idea is to use Alledia's SEO forums. As I mentioned, he is very good at Joomla! SEO questions.



Thanks for your very informative post. A way to automatically match the old URLs to the new URLs and generate 301 redirect statements would be helpful. Have you looked at the Migrator code? Could that be adapted?

HatonCat indicated he would talk to Alledia. Steve Burge offers many Joomla! SEO services. Maybe this is a service he might be willing to provide for customers for J! v 1.5 migration? I'll give him a shout, too.

So, that's a couple more productive ideas to consider. Thank you.
Amy

Thanks for people pointing out this post to me. I understand where hatoncat's concerns, but my thoughts would be these;

1) The Joomla team have been under quite a lot of community pressure to step up the SEO features in 1.5 and it looks like RC2 will bring a lot of those improvements. The URLs in 1.5 were a particular source of friction with a thread running on for pages and pages: http://forum.joomla.org/index.php/topic,427.0.html . Credit where credit is due. 1.5 shows a lot of improvements.

2) The Joomla team would need a legacy mode to account for both default URLs (index.php?) and default SEF URLs (content/view/211).

3) Yes - a legacy URL mode would be nice, but I wonder if this request is one too many? If it takes a considerable amount of development work (as I suspect it might), would the advantages be worth the effort? Would keeping the legacy URLs be considerably better than implementing redirects from old to new? If Joomla websites were particularly old that might be a serious issue, but by definition most Joomla websites are 6 months to a year old.

4) Is this a situation that is better handled by interested people pooling their money and hiring a developer, or by asking the core team to do more? I suspect the former.

As to point #3, it is a cost-benefit analysis. However, I think people are underestimating the cost. The cost has to be measured on two ends: Those that won't migrate because they know the damaging effect on their site would be great. And, on the other end of the equation... the cost of all the sites that unknowingly migrate and lose their backlinks. That end will cause great frustration, because nobody (at this point) is warned during migration that backlinks will break.

I think when you add up both costs, it should become a priority for Legacy Backlinks to be implemented in the initial 1.5.0 release.

As to point #4, I would love to throw money at the problem. But, GPL rears its ugly head. If some developers raise the money, they are essentially "buying" that code. The developer is then motivated to "sell" the code to more people (since the initial buy-in will probably not represent all sites wanting to migrate Legacy Backinks). That then violates Joomla's stance on GPL... since only the people that ponied up would be able to take advantage of the workaround.

Obviously, that's not viable, since most Joomla site admins (when informed) would probably be willing to pay to migrate their Legacy Backinks.

So, I think there are two solutions, one you sorta brought up. Again, core members writing this in would be the optimal solution. However, offering up a bounty to get the job done would also be potentially an effective solution. Once the bounty has been filled, that process can then be integrated back into Joomla itself (either as core code, or in the case of a not-so-elegant htaccess generator, into com_migrator).

In other words, I still would like those core team members working on Joomla 1.5 SEF to speak up (so far, they've been silent on this thread), and see where we can go from there. I think a bounty is a worst-case scenario option, since there's no guarantee it will be finished in time to be integrated into 1.5.0's release.

When it comes to chipping in to get this done:

I’m racking my brains – there’s a sort of reverse auction site available where a developer sets a price to get something done and everyone chips into an Escrow account. When the target amount is met the developer gets going and the result is released as GPL.

I seem to remember a couple of Virtuemart addons were done this way. Anyone remember the name of this system?

Hey!

I have been thinking about this issue and trying to devise a way to solve it.

How much money do you think could be raised to contribute to such a project?

When you say:


How many are you really talking here?  I'm sure there are more than just the two of you, but so far it has only really been the two of you...

Is it important that the links be redirected to the new SEO versions?  Or can, for example:
http://127.0.0.1/j10/content/view/9/33/ be converted to:
http://127.0.0.1/j15mig/index.php?optio ... &Itemid=33
be done?

Ian

No one is being "silent" regarding this request. Joomla! v 1.5 has been feature frozen for awhile. We are at RC 1 hoping to go to RC 2 this month. They are not certain if RC3 will be required, yet. This release is being steadied for blast off. So, changes to Joomla! v 1.5 are highly unlikely for any request. In Wilco's last blog, he said they would begin talking about v 1.6 soon.  That is when they will hear our requests for core changes and they have already said they aren't ready for that *quite* yet.

I wanted to mention that Johan built J! v 1.5 SEF URLs with additional system plug-in capability. If you have used WordPress or Drupal, you know that there are options for forming your URL. Johan has this foundation in place for 3rd party developers to exploit. Could that be *a way* this is resolved? It might be worth contacting your existing SEF URL third party developer and see if they plan to take advantage of this capability - or if they have studied it, yet. Perhaps, eventually, this will all be naturally resolved with the work already accomplished in the core, given a simple plugin.

But, there is a very good solution to this problem that has no cost and no work for anyone . Just don't migrate to Joomla! v 1.5.

I wonder how completely has J! v 1.5 been assessed for your site? Are all other extensions you require, in place and tested? How well did the migration tool work for your data? What did you think about performance? Are the URLs the only remaining issue you have to solve? If so, it might make sense to consider migrating and solving this one problem. Otherwise, there might be many other issues, as well.

Those who want to migrate to v 1.5 now, will have new sites with minimal need for many (or any) extensions, or are willing and able to build or pay for their own extensions There is just no reason in the world that anyone *has* to go to v 1.5. Those who need more "out of the box" solutions will want to wait until those solutions have been made available.

Thanks,
Amy

I'm not sure if you were referring to my post here, Amy.  I was just wondering whether it was worth my time to generate a fix or not.  There is a common problem in the forums that people will say 'everyone wants this', but it is often impossible to tell how many people this is really referring to.


This is indeed easy to solve with a simple plugin.  I decided to do some research into how difficult this problem would actually be to fix.  I ended up developing a plugin that so far handles the content component with the exception of content archives.  The plugin basically detects a legacy URL by checking to see if the URL starts with content or if option is com_content and the task variable is set.  If a legacy URL is detected, the URL is decoded, converted to the 1.5 format, and run through JRoute to produce the new URL depending on the 1.5 URL settings.

Then, the plugin will issue a 301 header redirecting to the new address.

It needs more extensive testing, but it seems to work as a POC.  If there is enough interest, I might look into handling other components.

If some wanted to help, what would be of most benefit would be an exhaustive list of the components that need to be handled and how the transformations are done.  I'm sure if there is interest, it should be easier for this information to be provided.  Ideally, what would be provided would be the 1.0 SEF URL, the 1.0 non-SEF URL, and the 1.5 non-SEF URL.

Ian

No, no, no! My apologies, Ian.    Not in the least! I was responding while you did - and I didn't quote hatoncat. This is the specific comment I was responding to:



My point is - the core has *already* communicated that J! v 1.5 is feature frozen - we are at RC 1 - there will be no more changes - this was announced some time ago. Soon, v 1.6 will be discussed. That's when the core will request our new ideas. Lots of people make requests but the core doesn't individually respond to each person. They communicate via their blogs, etc., as to where the process is at.

Again, my apologies, Ian.



Yes, I understand. You are, indeed, one with the skill and knowledge of v 1.5 to develop a solution. But, you need to know "real" demand. Or, better yet, commitment.

From hatoncat's comments, it sounds like he is willing to do so:



So, Ian, will the GPL issue stop you? Or, would you be willing to liberate your solution if you were paid for development? The other concern I would have, like you are saying, is getting good specs. Hearing the change must be in core has no meaning.

Thanks,
Amy

I will contribute the code once it is complete.  I would ask that those who benefit would consider writing a tutorial documenting something useful they have learned about something that is currently lacking in the documentation concerning Joomla! 1.5.  This could be user documentation, or developer documentation.  Documentation detailing the ins and outs of the migration process would also be really helpful.

Ian

Two have already indicated they will pay for a solution - that this problem is that critical. From the sounds of it, there are likely others. You should be paid for your code, too. I think that is the point hatoncat was making about the GPL. Developers *should* be paid - and he's right. So, this could be a model of that success, how a bounty could be used to generate GPL'd code, and how end users show commitment to pay developers.

I understand documentation would be helpful, too. But, during this transition period, a developer success at making money with GPL code would be excellent. I think we all agree on that!

Thanks, Ian, for all you do always,
Amy

Amy has convinced me that perhaps I shouldn't be so quick to give my work away.  I'm contributed a great deal already, both in the form of documentation, and in the form of other extensions.  But this case is unique.  I would never have considered taking this on had it not been made such a big deal out of by those in this thread.

I'm resistant to charge for the work I do related to Joomla!  Not because I believe that software developers shouldn't get paid, but because I do what I do because I enjoy it, and I don't want money to get in the way of it.  It has only been recently that I have taken on any paid Joomla! work, and even that was only because more of a time commitment was required, and it was for an organization that I wanted to support, so I have done it at a rate that was much lower than I believe most in the business would charge.

Both my wife and I work for non-profit organizations.  We don't make loads and loads of money - and this was a choice we made, and a choice we do not regret.

All this being said, it sounds like there are people who stand to make big money as a result of this work that I have done.  I don't want to ask for a lot, but some financial remuneration would be really helpful, especially as my wife and I look forward to having our first child, and the realities of the costs involved start to sink in.

I would certainly be willing to negotiate a fair price with those who would be willing to sponsor me to release this code under the GPL.  Those who are interested should contact me via PM.

Ian

And what about the links between different pages of the site?

When url scheme changes, these links become dead as well. Is this problem solved? (I believe that some plugin-based solution should be possible, and probably an .htaccess-based one which will also account for the inward links.)

PM ianmac. That's what I would do!

I'd like to offer a small insight into this issue:

It has nothing to do with upgrading to 1.5....

This is all about SEF and your URL. You would have the same problem (and I have said this before) if you were using one SEF version, say SEF advance, and decided to switch to another, say OpenSEF. Or had no SEF, then turned on Joomla default. They are all different.

Right now, I would think most people are using a 3rd party SEF. Once you start using it you are somewhat trapped, because of the whole backlink issue described by the original poster. That makes this issue nothing to do with the core code, few people use it anyway.

If the 3rd party dev wants to keep his customer base S/HE will develop a 1.5 extension that duplicates their own unique way of building the URL. All his/her customers that are using the extension will use it again, or go through 301 hell....

To put that in perspective, the reality is that the core code would need legacy/redirection modes with all major 3rd party SEF extensions, having legacy/redirection for the core SEF is moot, few use it. That's a daft situation imho....

I don't know that I would call that a 'small' insight by any means, Barry, and it makes perfect sense in my opinion.  Thanks for weighing in; that's why I pointed this post out to you from your blog comments.

This is really not so much a core issue as it is a "3PD working WITH the core changes" issue.  So it's either dependant on the multiple SEF vendors to get involved or the individuals with the most at stake (the users with financial stakes attached to their backlinks) to move things along.

I don't know about how many people use the core SEF, but I'm one of those few guys who don't use SEF at all. So I would appreciate very much if Joomla! team provided the solution for these two groups of people.

I agree with compass on the technical side of things, as well as with statement that it's 3rd party developers' responsibility to provide an upgrade to 1.5.

I used to be one of 'those guys'  , but add on-s like JoomSEF made it very easy for me to get into it on the 1.0.x platform; you might want to look into it. 

J! 1.0x core SEF was always a 'black box' affair to me; more often then not I would give up trying to make it work.  But 1.5 core SEF worked for me the first time I enabled it and I think that's going to make it accessible to a LOT of users who had issues with the older SEF.

Hm. I'm not certain everyone is reading this thread closely.

A solution has been proposed. Ian indicated the problem can be resolved with a custom plugin. It would be the same solution regardless of SEF URL tool - or none. He would likely build it if a group wanted to hire the work done. Or, anyone else can build it, too. The Joomla! core team cannot be expected to resolve all of our problems.

Perhaps the original posters and others might want to pool resources, contact Ian and see what it might cost to have him do the work. I am certain it will not come anywhere close to the tens of thousands of dollars earlier reported.

Amy

Actually, it might have been free until you opened your mouth...  



Just kiddin Amy! And Ian, I agree that if someone has already attached a financial cost to the problem, there should be renumeration for the solution...I just couldn't resist a joke at Amy's expense...

It was already proposed in this thread. You only had to read! 

Open source isn't about free, as in beer, just free, as in liberty.  Many of us believe it is important to pay open source developers.  In fact, that was a lot of the ranting in this thread.  So, hopefully, people will demonstrate this conviction.

Amy

Please Amy, no soap box required - I tried to make it abundantly clear that I was attempting to inject a little bit of levity, okay?

Hey! I winked, too! 

No worries!
Amy

Sorry, I guess I mistook it for a glint off your glasses 

Cheers!

The problem is that a short term solution can be done in terms of hacks and plugins. I agree, I have looked in to the code, along with some of the other SEF component communities. It can be solved as a migration plugin sort of thing without too much work.  But, as mentioned above the problem are:

1. All the various SEF url components that have been used because Mambo did not do it right from the start.

2. All the third party components also implement different urls. As already pointed out the list of components with different url rendering makes it fairly serious tasks to capture them all.

3. we have different servers at work. The two big ones to handle are Apache and Windows. This is seems to not be too bad a problem in terms of kicking out 301 redirect solution for each.

The problem with the plugin solution is that it is exactly what has caused the mess to start with. Everyone has implemented their own way of rendering and controlling urls, when there should be a built in core feature of the system to allow everyone to control the way their urls are rendered in Joomla, and easily redirect them from one page to the next.  That way, such as in this case of the 1.12 to 1.5 migration, going forward links will convert with joomla as joomla progresses. 

The urls are something that should be both automatically generated, but also something that needs to be tweaked for SEO purposes on an individual page by page case. In the same way we have a box to add a meta description for a page, we need a box to be able edit the pages url. Yes, there are lots of plugins and components that do that, but again that is exactly why the url mess exist in the first place.

What happens when another big migration comes along?  It is a problem that needs a core solution for the long-term.

Might also point out from an SEO standpoint it can be just as big a problem to move all urls at once on a site with 301 redirects, as to not redirect properly. Sites intent on keeping their ranking should be stepped a few at a time to avoid dropping out of Google for anywhere from a month to 6 months as Google updates and redirects. 

So, URL control is what we are after as a core feature in the same way users want to control the title of a page or a meta description.  They are basic and important functions of any website and something that you should be able to control with joomla just as easily.

I think one of the best arguments for sticking these functions  in the core are made here by cenc; namely standardization.  That's not to say that this can't be done by other working groups and adopted into the core, but IMHO at some point this function should be partof the standardized API so that going forward we can have some epxecations of uniformity.

It would be great to have this, and many more things in core. Who knows? Maybe those volunteering those efforts will decide that is what they want to do. For now, however, thankfully, it sounds like we know other options that can help. Those options only need to be pursued. I think this is a case where the architecture - the plugins - are evidence of how flexible the core will be for many varied needs. That is a good sign!

Thanks!
Amy

Some have drawn connections between Joomla 1.0's SEF and OpenSEF, arguing that Joomla 1.0 SEF should be left in the dust since OpenSEF won't have a clear migration to Joomla 1.5.

I disagree with this line of thought for one clear reason: Joomla administrators were under the impression that Joomla 1.0 SEF would be supported. In fact, they still are... since most are unaware that their backlinks will fry when they migrate to Joomla 1.5 (and I don't think anyone is arguing that most admins are unaware of that right now, it certainly isn't documented in any part of Joomla 1.5's release notes).

I personally opted against OpenSEF, or any other third-party SEF component because of the fear that they would become dead-end solutions. Never in my darkest of fears did I think that Joomla 1.0's own official SEF system would wind up in the same boat. I think that 99% of Joomla admins out there, would agree with that.

Joomla 1.5 finally has great SEF, but Joomla 1.0 SEF users certainly have a good argument that their SEF is official and supported, and thus should be migrated. As to OpenSEF... unfortunately, Joomla admins that chose that route, chose one that is not officially supported. Either the community over at OpenSEF will make their own migration, or those admins will (unfortunately) have to chose between migrating or retaining their backlinks.

One thing that I do fear though is that the OpenSEF community, being much more savvy as to matters of SEO, will come up with their own solution, and those of us that stayed loyal to Joomla code will be left in the dust with dead backlinks, and ruined PageRanks.

hatoncat -

What do you want?

What are you willing to do to make that happen?

Thanks for clarifying your post!
Amy

Amy, I regret that you continue to consider my posts as being unproductive. I hope we can move past that mindset. My previous response was to a post that tried to argue that official, supported Joomla solutions were in the same boat with unsupported solutions. I disagreed, and provided cogent reasons why.

What I want is to see Joomla succeed as a CMS. I see this as a flaw that could severely damage this goal. I want Joomla 1.5 to have a documented, full migration of legacy backlinks for those that embraced and stayed with Joomla 1.0 SEF through thick, and in recent years, thin.

What am I willing to do? Well, let's start with what I've done so far. First, I've raised awareness about this issue... there was not one line of discussion about this matter, and now people are talking about it. Second, I've been in contact with well over a dozen people in public and private to see what the best options are for moving forward. I've brought some potential solutions to the table, others have provided some other solutions. Unfortunately, the leaders in Joomla SEF haven't commented on any of them.

My post last week calling out for core team members to comment has gone unanswered. They are being silent, despite your assertion that they are working on Joomla right now, this issue needs some form of redress from them. I am trying to get their comment... hopefully you can assist in bringing those that developed Joomla 1.5's SEF (and 1.0's SEF for that matter) into this thread to further the discussion.  There are too many options on the table that have not been evaluated by those that can directly intervene... and possibly quickly resolve this.

Edit: I know you have said that the "core has spoken" because RC1 is frozen. I consider that not logical. If they truly believe that, they should say it here. That is like saying that a critical bug in Windows that blocks half the users from installing... won't get fixed until the next service pack. Understand, I do not see the lack of legacy backlinks to be a missing feature, I see it as a flaw. You are welcomed to disagree, but I don't see anything productive stemming from that debate.

So Amy, I hope you can help to get those people to comment. I don't think my opinion is well-rounded enough to provide a best-solution until the folks that authored Joomla SEF respond to this situation. I will continue to work to get them to do so, and that is what I am willing to do right now.

Hat -

OK. Now I who you are now. I knew I had "seen" you around, but it's been awhile and I didn't immediately remember you from this thread. Your post was honestly confusing and, no offense, but it was rambling. That's why I called for clarity.

Last time we talked, we found someone who can write a plugin for this purpose. He is very knowledgeable and willing to do so for hire. Did you contact him? As you rally support to have this work completed, do you also ask people if they are willing to contribute to the solution? That would be *very* productive!

Here's a rough analogy on why it's not necessarily as helpful just to find support. If I walked door to door in my community and asked people to sign a petition ordering the city to stop taxing us, I'd get a LOT of support. But, if I also asked that they be a part of generating the revenue in another way, support would drop.

The community requires contributors and we cannot demand volunteers do more and more work for us, no matter how passionately we believe THEY should. As I said earlier, Joomla! v 1.5 has been feature frozen for some time. Happily, it turns out they created this nifty plugin environment for us that ianmac believes would be useful to solve this problem. He agreed to do so for a price. The next step is to ask him.

Please try that if you haven't already. Helping fund that work would be contributing to the community - a contribution many would appreciate. It would also help us understand that we can financially support our developers, even with the GPL.

Please give it a chance.
Amy