This discussion began in a post to the Lyceum Users email list.
Lyceum users,
As many of you have surely noticed, Lyceum development has been essentially nonexistent for the last year and a half. The last point release was a bugfix in June 2008, which is still based on the WordPress 2.0.x branch.
For a lot of reasons, I have not had the time to devote to Lyceum that I used to. Additionally, Lyceum’s primary competitor, WordPress MU, has grown into a respectable and consistently-maintained project. While some users still tell me that they find Lyceum easier to install and in various ways more intuitive than MU, it has more-or-less feature parity with Lyceum…
…except, of course, for the database schema. Lyceum’s schema is normalized, MU’s is, roughly speaking, sharded. For massive installations such as WordPress.com, a sharded schema is preferable. A normalized schema is preferable for everyone else. A normalized schema offers advantages in performance, ease of upgrades and administration, and vast opportunities for developing plugins and features which can access the entire set of data in the installation.
“Show me the last 10 posts across all blogs” is impossible in MU without maintaining and accessing a separate, redundant index table at the application layer. In Lyceum, it is trivial and fast, even for thousands of blogs.
Regardless of these advantages, the effort of maintaining Lyceum must be matched by demand. So I find myself at a crossroads, deciding whether to pick the project back up, or let MU be the only option.
If I choose to resume Lyceum development, here are the two initial phases of development:
Phase 1 - simply bring Lyceum up to the current version of WordPress
I have two options for doing this. The first is to continue developing Lyceum as a branch of WordPress. The second, and more likely option, is to start from scratch using the WordPress MU codebase, keeping all of its features and merely normalizing the schema.
Either way, I could get Lyceum back on its feet in 2-4 months of development. And don’t worry: even though the codebase would be wildly changed, there would be a fully-supported upgrade path from previous versions of Lyceum.
After getting the codebase contemporary with WordPress, applying future WordPress updates to Lyceum would be a non-trivial, but relatively straightforward process. I estimate it would take me an upper bounds of 5 hours of work per week.
Phase 2 - add Lyceum-specific features
There are two main types of Lyceum-specific features that I would like to develop. The first are features which simply leverage the normalized schema — template functions, hooks, utility pages, site-wide feeds… all accessing site-wide content and objects in a way that is easy for template and plugin developers to use without having to worry about the complicated denormalized meta-tables of MU, or the mind-bending (yet fully index-covered) four-table JOINS of Lyceum. (Or any sql at all.)
The second are features which evolve Lyceum into a social platform, connecting content, users, and their online publishing presence in an intuitive and ergonomic way.
I could go on and on about even more things I would like to add to Lyceum — the main theme being seamless integration with other platforms, frameworks, and architectures.
I’ll stop here and let you folks respond.
- What value does Lyceum hold for you today?
- Did you chose Lyceum over WordPress MU mainly because of the schema?
- What do you like and dislike about Lyceum?
- Do you think Lyceum development should continue?
Thanks for any and all feedback, and thanks for using Lyceum!
John
41 Comments so far
Leave a comment
I installed WPMU a couple of years ago at our community site for almost four days before it bugged out, then I discovered Lyceum. I think that it was at around .4x or thereabouts, and the advantages seemed tremendous to me at the time. Sadly, I never did complete the install, and settled on a profile blog instead, which is woefully inferior as far as features, but it’s easy.
We’re running a sort of integrated Gallery2, Joomla! and Simple Machines Forums operation. I decided to check on Lyceum again this morning and discovered this post.
Please consider continuing development, there should be a huge and hungry market, especially in the integrated CMS and forums communities. It might be easy to find a few coders for each specific platform willing to help develop a “bridge” to wrap and integrate the user tables, set up menu links and the rest.
In the meantime, I guess I’ll check out the 1.03 build, and see how that goes. Anything to add value to the overall experience.
Thanks for your efforts, John.
By Richmedia on 03.12.09 3:36 am | Permalink
Please continue with this good work.
I want to be able to post “the last 10 posts” and so on.
Besides, competition is good.
I am in the process of setting Lyceum up on my site… I hope it goes well.
By diane on 03.14.09 10:55 am | Permalink
I’m a web designer/developer, and I often use WordPress as a simple CMS for clients. Sometimes I need a multi-blog platform, and an alternative to WordPress MU with a “normalized” schema would be preferable.
By Jason Loftis on 03.16.09 7:43 pm | Permalink
what? lyceum imperiled?
There are a few of us, of whom there exists no record other than an entry on an IP in a download log, who would be devastated.
we run private servers, on the web, just-for-fun. Without actually useful programs like Lyceum that work, correctly, without using all system resources on a P3 at 1ghz, life would be nothing more than hacking other people’s naive sql installation procedures and correcting CSS errors. where’s the fun in that?
By maxnort on 03.17.09 7:15 am | Permalink
I’m sad to hear that you’re considering not developing Lyceum any further, as it seems to be exactly the solution I am looking for. WPMU doesn’t adequately foster sitewide integration with individual blog authoring.
By Daria on 03.18.09 2:35 pm | Permalink
Looking for a system that could handle 100 or so blogs I downloaded and played with Wordpress MU today. Now, I am into performace and I found that the plain Wordpress MU install used 9.8MB of RAM and included 62 files on each load. That just about made me sick.
But it gets better, when moving on to install BuddyPress to extend the WP MU to do a networked site - the includes went up to 132 files (!) and 16MB per-page load. Wow, so you basically have to be rich enough to purchase your own datacenter to run this system huh?
After building MicroMVC I am used to small sites like my blog being able to cache pages and limit memory usage to 180KB and system load times to 0.008sec. Wordpress is just way to big - which is probably one reason they didn’t mind releasing it as only wordpress.com could handle that type of load (so they can pick up the users that the WP MU setups find they can’t handle).
Anyway, I guess WP MU + BuddyPress is my only choice for now, but if you would like help “slimming” down the code base of the extra junk - let me know. I will have to do that just to run it anyway.
By David on 03.19.09 6:25 pm | Permalink
yeah I agree with “David”
By makaveli-mk on 04.01.09 4:34 am | Permalink
I’m currently hosting 20+ wordpress blogs (and growing, slowly), with the debian single-code-multiple-configfile method, so there is only one instance of wp.
I wanted a better way, so I installed WordPress-mu on a new vm, got everything up and running, and then discovered that it only supports subdomains of ONE domain!
WTF?!
What good is WPMU if I can’t set it up to host different, independent domain names?!
Sheesh.
If Lyceum supports independent domains, then yes, I’m enthusiastically in favor of continuing - there is a real need.
In the meantime, back to supporting 20 (oops, now 23) separate dbs, 23 separate admin upgrades every time WP releases a new version, etc.
By Joe on 04.02.09 1:48 pm | Permalink
I am currently trying to find a way to consolidate management of “staff blogs” for a school newspaper website. I’ve explored CMS Drupal and started to read up on Wordpress MU, but it seems like Lyceum could actually be the easiest solution. I’m not positive it’s what I would need, but if development resumed, I would love to check it out and see if it works for our needs.
By Chris on 04.04.09 4:11 pm | Permalink
I just browsed around for another solution for WP MU when then I found Lyceum link on Wikipedia.
Well, I think lyceum development should really continued.
It’ll be the answer for many people who looks for reliable multi blogging platform ;)
By KaiToU KiD on 04.04.09 5:01 pm | Permalink
Consider this another vote for continuing; I’m in the process of deciding how to do a multi-blog install now, and Lyceum seems MUCH more reasonable to me that Wordpress MU. I’m a database admin by trade, so I appreciate the pros and cons of the schema differences…your solution is much preferred to me. There are VERY edge-case situations where a site might scale to a point where sharding is required, but your solution is far better for 90+% of installs, IMO.
Here’s to hoping for a long and successful future for Lyceum!
By Ken Kennedy on 04.04.09 9:28 pm | Permalink
HI,
I have come here after a long search for multiple websites on independent domains using a single wordpress install. I have banged my head over WPMU for 4 days and nothing came out of it. Please consider this another vote to continue development.
By icubyx on 04.05.09 2:20 am | Permalink
WPMU’s achilles heal is lyceum’s strength: the approach to databases architecture.
In agreement with @david and @joe’s comments above.
Install a single instance of the codebase and re-use it over whatever domains and/or sub-domains I wish should be easy enough in this open source day and age. Sadly, it’s not the case…yet.
My question is simply, can lyceum be brought up to date to allow all the buddypress plugins to work, since buddypress is basically ONLY a set of wpmu plugins, hence lyceum + buddypress = an easy on the server resources social network ?
If that could be done, they there is the core reason for lyceum to exist and not to be abandoned.
By HiMY SYeD on 04.12.09 11:21 am | Permalink
JJB,
Don’t let lyceum die- maybe some of the commentators to this post can be roped in to help out with whatever direction everyone thinks is best.
thats’ just my $.02
BTW, hope all is well at your end.
’till later buddy…
By Sayan on 04.13.09 12:10 am | Permalink
I would really like to see the project continue. If you’re open to it, I’d like to contribute as a coder.
By Jason on 04.15.09 8:14 pm | Permalink
[...] headache I have with WPMU is performance. There are some disconcerting reports on the amount of resources WPMU consumes per page load. For wordpress.com this is not an issue, since they run their blog farm on dedicated servers (and a [...]
By WPMU Blues | Hackadelic on 04.19.09 11:00 am | Permalink
I think WP(MU) owe their popularity to factors other than good software architecture. So I personally would find it a pity if a(nother) system superior in this regard would vanish. So take this as a vote pro Lyceum continuation.
Having said that, I notice that it is unclear to me why Lyceum is tied to WP (beyond the fact that it started sort of as a WP fork), if the two are basically incompatible at the data level. What is the benefit from it for the Lyceum project, and what is the benefit for Lyceum users?
If there is real benefit in it (which I don’t doubt is the case, even if I don’t immediately see it), then synchronization with new WP releases should be largely automated.
Do you think that is feasible?
By Hackadelic on 04.19.09 11:41 am | Permalink
2continue
By michael on 04.22.09 9:35 pm | Permalink
Hey,
I was waiting for an update on Lyceum for a while. I have a set of 5 blogs i would love to have a single admin for, but sadly, have never been anpe to get either WPMU or Lyceum to get started on.
I just downloaded WPMU, and came here to download Lyceum to see which is better, when i read your post.
Do continue, maybe post on the WP forums, am sure you will find coders and the project will be running earlier than you predict!
By Jatin on 04.25.09 7:58 pm | Permalink
Just a thought, but if you do continue and decide to rebuild based on the latest version of WPMU, why not look at creating a plugin or some simple modification that would normalize the WPMU db structure? It seems to me that if your system is simply a copy of WPMU with a different db structure, this would be much easier on you and also be beneficial to many more people as a whole.
However, if you do decide to continue your own project, try to build something that is as flexible as other CMA programs out there but with the simplistic and user friendly access of WP. I love WP, but the one thing that bugs the hell out of me is the developer’s reluctance to expand WP into a full fledged CMS rather than trying to keep it as a simple blogging system.
Good luck in your decisions…
By Ed on 04.27.09 9:34 am | Permalink
@joe WordPress MU does support domain mapping with this plugin:
http://wordpress.org/extend/plugins/wordpress-mu-domain-mapping/
By Lyceum Website Admin on 04.27.09 5:35 pm | Permalink
@HACKADELIC the goal is plugin and theme compatibility
By test on 04.27.09 6:11 pm | Permalink
@ED lyceum contained in a plugin… a tantalizing idea, but i don’t think something that would work in the long term. regarding full-blown CMS… I think WordPress’ focus is one of its strengths, for the time being. The pressure is certainly on, though.
By test on 04.27.09 6:13 pm | Permalink
This project seems to be exactly what I’m looking for. In fact I got here from the WP site where they list you guys as a multi blog alternative and they even list you before their own WPMU.
Here is the link
http://codex.wordpress.org/Installing_Multiple_Blogs#Alternative_WordPress_Projects
So I’m hoping you keep this project alive, at least, as you mention, synchronized with the official WP releases. Keep up the good work.
By salsaman on 05.02.09 12:29 pm | Permalink
Hi,
I’ve used lyceum since v1.0.3. We’re hosting ~1400 blogs (surely enough only minority of these are regulary updated), and lyceum offered a much better solution than WPMU at the time. I’ve made many various hacks for admin part, such as front page (portal.php) posts selection from all the blogs, automatic theme selection for some user groups (editors), global archive page/calendar, used memcached to minimize db load, using custom centralized authentication, blog posts generation to static files etc. The main issue now is that it’s kind of outdated (wp 2.0 interface), and there still are some of the bugs I haven’t fixed myself, emerging from my hacks. Yet this is a good system (better than WPMU since I need some custom hacks from time to time) and I’d like to see the future for it.
Thanks for your work and hopefully we’ll se v2.
Regards,
d.
By d. on 05.22.09 3:03 am | Permalink
I believe healthy competition is good for the open-source community. So, my vote is for lyceum to continue.
It would be nice to see lyceum, or some other WPMU fork, evolve in a different way than the current incarnation of WPMU. With social networking playing a more important, dominant role in CMS’s–take Automattic’s funding of the BuddyPress plugin as an example–the old blog-centric paradigm needs to end.
Instead, the foundation should focus on users. WPMU’s core is so tied to the blog as king that it can be very hard to work around that concept when trying to build certain custom BuddyPress components. The paradigm needs to be user-centric, not blog-centric.
Therefore, I would like to see the following:
1. The foundation should be the user, not the blog. After all, blogs are owned by users, not the other way around.
2. The core should be truly object-oriented. Currently, WPMU is some mishmash of object-oriented PHP with non-OO PHP.
3. The user should be seen as a super object that owns and controls all other objects. The Site Admin should be seen as the super user that has ultimate control over all users on the install. A blog, then, is just another user-owned and created object. Currently the Admin dashboard is organized around the blog, not the user.
4. When installing, there should be an option to choose normalized versus sharded DB schema.
5. There needs to be a true database abstraction layer that allows for the use of competing OS DB servers. There should be an easy way to choose between using MySQL and PostgreSQL for instance. WordPress’s wpdb class is just not adequate anymore.
6. Plugin conflicts caused by activation sequence order are becoming more of an issue–especially with the sophisticated BuddyPress plugin. More experienced plugin devs do add code to check for the installation and activation of certain core plugins before their plugin is activated, but there is a better way to do this. There should be reserved blocks of numbers set aside for key components (plugins). These numbers would be lower in the activation sequence, therefore gaining higher priority. So, BuddyPress for instance would always be assigned a reserved activation sequence number that is lower than any 3rd-party plugin that is developed to run on top of BuddyPress. The other way to do this is parent-child dependancies. All BuddyPress plugins would be a child to the parent component.
Whereas I am thankful for an appreciative of the contributions that the WordPress team has made to the open source community, I also am a proponent of healthy competition. I believe that the underlying code foundation should evolve to offer a more flexible way of doing things. Perhaps Automattic will move in this direction!
By Jeff Sayre on 06.12.09 10:48 am | Permalink
John,
I’m afraid this will be a fairly uninformed comment for you since I am still on this side of WP installations. However, I do own multiple websites and want the ability to set up a single WP system that will permit multiple blogs and Lyceum sounds simpler and cleaner. Normalized would work for what I have in mind. Competition helps and is healthy and if you can manage it, keep on keeping on, please.
By Channing Grigsby on 06.24.09 9:18 am | Permalink
I’ve just started using WP and really want to create a magazine style blog with 3 different blogs running under 1 main site. WPMU sounds like it isn’t the place for me as it is still very buggy?!
Keep up the good work
By JAsper on 07.04.09 2:19 pm | Permalink
DO IT PLEASEEE!
WE NEED YOU!
By Anonymous on 07.16.09 12:38 pm | Permalink
Vamos amigo continua con esto; te sigo desde hace 2 aƱos y siempre estoy a la espera de una nueva version.
We continued with this friend, I keep you from 2 years ago and I am always awaiting the new version.
By MasterChowl on 07.19.09 1:58 pm | Permalink
We have been searching for a simple multi-blog platform, and Lyceum seems to fit the bill. At least, from what we have read so far, more preferable than MU for our situation. I hope you continue, and I hope you commit to that path soon, as it would surely help us decide. Thanks for all the great work so far.
By Scott T on 07.31.09 2:33 am | Permalink
I am looking for exactly this kind of solution! I want to have about 20 wordpress blogs on sub domains, and put all latest posts on the main site. Unfortunately this seemed to be really hard to do in a convenient way in wordpress, until I found this project! please do continue, I don’t believe I’m the only one looking for this kind of solution..
By Steini on 08.04.09 1:44 pm | Permalink
I definitely have to give my vote for continued development. I’ve been maintaining a fairly large MU install and I continue to be annoyed by all the limitations created by the database schema. What’s the point of a blog network if you can’t easily utilize the data from all the blogs? We have all sorts of stability issues and I would love a better alternative.
I think the best way to proceed would be to start over with the current MU codebase. It’s probably nearly impossible, but if Lyceum could run as a plugin for MU that would be incredible.
I have quite a bit of experience with WP and WPMU so I’d be willing to help in any way I could.
By Aaron on 08.31.09 7:47 pm | Permalink
waiting for the new version
By adil on 09.01.09 9:24 am | Permalink
Being new to the Blogging world, after doing research, to learn, which is the best and easiest to install and use. Mu does not seem to be for the faint at heart, those not as code and programing savvy. That’s why the world needs more like you, that have the ability to take and improve on a concept. Which ever way, given time available that you have, that you choose, Lyceum and you are needed in the blogging world. Thanks for all your efforts and please keep it going.
By carey tucker on 09.04.09 8:07 am | Permalink
I think you should keep up going on it. It has many advantages in concept and in reality.
Thanks and go on
Regards
By Ali Jam on 09.09.09 1:23 pm | Permalink
OK, let me tell you a bit about MU:
- it’s great software
I hosted over then 14.000 blogs on MU and I’ve seen max scalability with my own eyes:
Quad core 4GB, RAIDII Dedicated machine dropped dead 2x week. No machine can handle 104.000+ mysql tables!!!!!
OK, like I said, great software, but I feel need to argue with you about saying that MU DB schema is better for large blog communities? Not exactly.
So, please, update this to current version.
I’ll make sure to recommend lyceum to anyone wishing to use multiblog WP features.
110.000 mysql tables just don’t make sense.
Cheers
By Romano on 09.26.09 5:11 pm | Permalink
I’d use Lyceum in a heart beat if it was under development. Thank you.
By Fima on 10.05.09 8:33 am | Permalink
i think your code is not clear:
1:install is diffcult
2:table design
3:urlrewrite is not clear .
4:never upgrate.
By code on 10.26.09 3:27 am | Permalink
I like Phase 2. If you need any design help contact me. I have used Lyceum in 2006. But moved onto focusing into an entirely different language and moved into developing with frameworks. I hope to help when I can.
By katz on 11.12.09 2:25 am | Permalink
I wish for the future development of Lyceum. All the best.
By UpadeshS on 11.24.09 5:57 am | Permalink
HTML allowed:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>