May 12, 2010 at 17:43Browse CMS by Category
In the world there are thousands of CMS for a variety of purposes, of very different quality, very different perspectives, cost, prevalence and so on. Seriously trying them all is unrealistic. Therefore, when I first got acquainted with the world of site engines, I had to choose at random. Below I will describe my impressions of getting to know these or those engines for various purposes. To some I will attach a brief description of the features, the impression of others consists only of peeking into the admin panel. These notes were compiled and edited for a long time, but now I decided that it is better to publish them in their current form than another six months to a year to edit a little bit without adding anything fundamentally new.
The advantage is given to free engines. Paid will be considered only for comparison or from hopelessness, i.e. if there are no free analogues. Also a limitation on technology: php. About the engines on the pearl and python, I no more than heard, on sharpe and Java I was dealing with self-written ones.
So are being considered
There are few comments, I am the moderator of the official forum and generally one of those who have been following this system from birth. Therefore, if anything, we will talk about MaxSite separately.
The system is written in a framework, therefore, you can add any functionality as you like. And the system itself on top of the framework provides a large number of service APIs. Therefore, there is nothing to be surprised that it will be found in other sections.
The structure and sharpening of the engine is initially a blog, but as an example of the same WordPress shows, this is not a limitation. Advantages: a good architecture and strong caching give good performance, ease of writing extensions - the whole system takes care of the service part, flexibility of settings - the conditions for displaying widgets, building links, constructing data types allow you to make sites very different in structure from blogs.
There are still some forks, someone else's heir, and so on. It seems that the neutrino is the heir to the mosquito, and the neutrino atomic editing is the fork of the neutrino. A CMS that does not use a database (and I needed it once). For simple business cards and blogs, it’ll work. Moreover, plugins can be written to it, and some of them are already available. There are publications feed, categories and tags, search (case-sensitive), templates, CNC. In atomics - in templates, you can control blocks (very convenient, actually).
Upd . Neutrino Atomic Edition will still be better, although even better see below.
I had to test the whole company: Neutrino Classic, Neutrino Atomic Edition, Nanote and Mosquito Blood Mary. Of all, their best impression with a significant margin was made by a mosquito. It all started with the installer - none of the above have it anymore. Further, absolutely all manipulations are done through the interface, and not manually. Everything is convenient and understandable. A minimalist file system, but for many cases it is enough. Plus the most friendly plugin creation documentation. And the possibilities are plus or minus all the same, maybe the mosquito is the most functional, and the nanote is the least functional. Blocks, tags, comments and everything about them, rss, cut, minimalistic file upload, etc. More about the engine .
One of the authors of this engine at one time created the Pikateka gallery, about which below. This is another engine that does not use a database. The engine on the files, but not the same as the above. As modules of this engine there is a forum (the most primitive), a catalog, ratings and even, it seems, a store. The engine also supports registration, user rights and all sorts of other things.
The problem is the same as with Picateca - the project died in 2007. Judging by the reviews, another problem, when used actively, indexes of the file database are damaged, i.e. Needs permanent backup. And I mention only then that sites that do not use databases are very well kept under version control systems.
The third Russian-language engine I met on Codeigniter. The first two are MaxSite and Cogear respectively. If MaxSite is basically blogging, a “wordpress killer”, and Kogir is a social network, then Image CMS, as you can see from a cursory glance, is more of a portal engine. At the moment, he is still sick with some childhood diseases, the first of which is few supplements. But otherwise it's a good engine, somewhat reminiscent of modx. The functionality is standard: category tree, tags, comments, widgets, photo insertion and gallery. Additional fields allow you to construct content types yourself. There is simple_cart. And the main advantage over the first two engines: multilingualism out of the box. True, I did not research it, but at first glance it was made better than in MaxSite.
Acquaintance began the year back in 2006 or 2007. Then the first branch lived and developed. The system was, as they say, not for the programmer. From the admin you can design absolutely everything. But at the same time, slowness was felt. One of the main drawbacks, but I still did not know that this drawback is the lack of Unicode. But at the same time there was a gallery (and not one), there was no way to organize CNC, multilingualism (buggy, but better than nothing), something else ...
Now I had to face again. Actual version 1.5.x. Unicode is. A bunch of already shitty plugins. CNC, forum (s?), Galleries, social network (s?), Store (s?) And so on. Slowness, in principle, remained, but still recommend the system to at least those who more or less know it and to whom it suits (or those who already have a working project on the jumble) - I agree. Surely caching can fix a lot.
As far as I remember, Russian-speaking communities competed among themselves. joom.ru, joomla.ru or something like that. Now there are considerably more of them. Which facilitates the search for information and answers to questions.
Upd. I had a chance to work with Jumla seriously. Impressions of the developer are absolutely obscene. The user experience is very favorable. You can design absolutely everything. But to choose from the same plugins the least load system, the least buggy and the one that will not be abandoned (or transferred to a paid basis) to the next version of the jumla ... Faster to write yourself (not for the jumble, of course). In general, if you have been familiar with dzhumla for at least a couple of years and know the picture and prospects of the plugins you need, then it’s good. And, yes, in addition you should have a powerful server that is not afraid to turn over CMS with several thousand files and monstrous requests such as fetching one record in a cycle.
Another problem is that a lot of the information that can be found relates to the jumel 1.0.x, while version 1.5.x bears the same name simply by mistake. Internally - this is a completely different system. And also presumably this year branch 1.6 should be released, which is already significantly different from 1.5.
Another update . I tried 1.6 beta 2. I twisted.
What I switched to, leaving Jumla - the wordpress. The admin of the WordPress after Jumla was the pinnacle of logic and simplicity. Moreover, then the branch about 2.0.x was relevant. Unicode was already supported then. Already immediately in the box were normal CNCs, customizable to the same. At the same time, CMS tossed and turned over FTP significantly faster than the dummy.
The more I learned about WordPress, the more I got into it. Plugin - a lot. This gallery and forum, and multilingualism, and all sorts of catalogs, and shops, and to hell with horns. You can do anything from WordPress. And with the WP MU assembly, you can also set multi-blog satellites.
But at the same time, as my knowledge improved and the versions of WordPress grew, I was more horrified by the growing gluttony. And limitations to some extent. On version 2.8.0, it was possible at certain times (when accessing the archive of a large site) to see 12,000 queries to the database. This is one of the most impressive, but not the only example of what happens when a project has no leader and the roadmap is determined by voting.
Again, this system can be recommended. But at the same time it is desirable to know and be able to tune to maximum performance. A big request to WordPress fans is not to arrange holivars here, because I express a subjective opinion, but confirmed by experience.
Another monster that can do anything, which has tons of plugins for all occasions.
In the minuses, the great complexity of this system. Somehow it happened that for a long time trying to study it, I did not get along with it. The mice cried, pricked, but continued to eat the cactus.
And in the minuses there is a serious gluttony, not less, but sometimes even greater than the previous two engines.
The main advantage of Drupal is a very developed and well thought out hook system, which allows you to redefine almost any event in the system. Plus two of the most popular modules, cck and views, which allow you to construct arbitrary data types and their display.
Either the first, or one of the first galleries that I met. There are some complaints about the interface, subjectively, I don’t like everything, but this is because the gallery is focused on multi-use, multi-album art and so on. User roles, user albums, limits, moderation and so on.
There are many translations, skins (tabular and artisanal), there are plugins. Integration with dummy and probably some other systems.
The branch I met with has been developing very sluggishly for several years, mainly bug and security fixes. The next branch has long been in svn, but only recently has grown to the status of beta and "official" download.
I would say that you can put it if the gallery is the main part of the site. Moreover, there is a plugin that implements the simplest cms, so as not to install any other engine for the sake of several pages. By the way, for the sake of several pages you can put neutrinos.
After at least four years of development, they posted version 1.5.x.
In contrast to coppermine, I really liked the picateca. Gallery is pretty simple. Easy fast and comfortable. Almost everything you need, almost nothing more. Instead of albums, tags. Multi-use is, but basic. The watermark in the code seems to be implemented, but it is not used anywhere by a separate function (by the class method), i.e. can not be enabled from the admin panel. And the main inconvenience: tag links use all the same tags, so for Russian ones links are terrible.
Abandoned at the end of 2006. No plugins are supported. The site has sunk into oblivion. If you merge from svn, then not the latest revision, but revision 179. By the way, there are bug fixes compared to the last archive.
I finished on her some little things. Porting a little bit to MaxSite. That just lacks a good gallery plugin.
Ps Now I found out that the author participated in the creation of ReloadCMS - a fairly large CMS that does not use a database.
Monster from the gallery class. The code is ten times more than that of the galleries of all others. There is integration with different CMS. But since I never needed such power, I never really set it.
Now I looked, I saw that in February, after several years of development, 3.0RC1 was released. It turns out that the third branch was rewritten in kohan. As a result, she thoroughly lost weight. Another interesting thing: the project started ten years ago. Perhaps this is the oldest project php-gallery of the remaining afloat.
If you need a gallery without a database. Two small ones, one fork of the other. Abandoned in ± 2005. Simple, with basic capabilities (except that there were no lightboxes at that time, but it is not difficult to adapt it).
Simpleviewer is another thing without a database. Flash viewer, it was interesting just in the absence of lightboxes and other special effects. Although if you need a flash, you can adapt now. The viewer itself is free, but there is a paid version for it with companions, the lack of a link for downloading, etc., and a free plug-in for wordpress. I’m thinking about how to adapt it to MaxSite. But not very active, because for large volumes of photographs it is of little use.
Recently I came across such a swatch. There appear in plugins: cms, slideshow, googlemaps, flwplayer. We are talking about support for Ajax, multilingualism, skins, password protection access to albums. Presented awards that the gallery received. Development began in 2005 and continues to this day. There are rewards.
She didn’t want to bet on the third Denver with a five hundred mistake - she was missing something.
The first instance that I met in this field. Although, I lie, the first was drupal. :) So, in terms of functionality, this is quite an excellent social network. Almost full analog of a habr. Personal and collective blogs, ratings, topic polls, topic links, invites. Unfortunately, the gallery is paid, and even the topic-photo report is paid.
And so the system is an excellent platform for social networks and adding the necessary functionality. The most technological of those that I saw, and so on. There is actually a very good OOP that reminds me very much of Java. It can replace a forum if users are not too conservative to this and if you work a little with a file.
As for the components: Smarty is used for display, the client part is mootools. For working with DbSimple data, for caching DklabCache (memcache is highly desirable). To search for Sphinx, which accordingly must be installed on the server. Well, and a number of libraries.
Actively developing (from September 2008 to June 2010 940 commits, two active developers and several dozen third-party add-ons undergoing quality control). I recommend it if the functionality is enough for your purposes. The advantage over Drupal and the company is that Drupal needs to be added to the necessary functionality.
Now about the shortcomings. There are two of them. Despite the good optimization, for a full visited social network requires a dedicated server, i.e. system requirements are not the smallest (see paragraph about memekes and sphinx). And to refine the site on LiveStreet, you need to lay certain resources of time and money or independent efforts. This is not a wordpress for you, for which there are free plugins and templates, and for whom it’s not enough, there are ready-made assemblies for all occasions.
The engine, in spite of which the "Living Street" was created. Some time ago I was in limbo. Recently sold, after which its development resumed. But it seems to be agonizing again. The choice between the two is a matter of personal preference. But personally, I did not really look closely. Therefore, I only know that for a live street, blogs on subdomains are implemented by a paid hack, and for a big one, it seems to be standard.
Upd . The new developer evaluates the kernel as functional, but generally artisanal. He is thinking about correspondence of the kernel from scratch and possibly on the framework - Zend or Kohana. But the development speed still lags behind the development of LiveStreet.
Another instance. Each branch is rewritten again.Soon a third should be released. But still the code is very scolded. I will not investigate more closely, because lacking tools without it.
Fork Express. More corrected, more secure, and in the upcoming release pretty rewritten. But even though the author claims that the project is alive, evidence of this is not visible. The latest release and the last commit in svn in the fall of 2008. Recent publications on this subject in the spring of 2009. Further developments in the public are not laid out. The author’s sites on his own engine are buggy. The project is clearly dead, unlike its predecessor.
Description rewritten . A social network written in Codeigniter. Of all the engines seen, the author of this has the most serious approach to documentation, appearance and all sorts of little things. In general, the author made a giant leap in writing this engine. A well-thought-out architecture allows you to do anything on Cogear if you have the knowledge, although the engine is geared to social networks. The disadvantage is that the entire author is drawn alone and obviously tired. Although it threatens to rewrite from scratch and not on CI. The plugin catalog now has a dozen add-ons and not a single topic, but writing extensions is very simple and fast due to the architecture and the hook system. Also, jambs sometimes happen, and the engine is completely not friendly with wine hosting. But the author reacts very quickly to bug reports.
Along with LiveStreet, it is a very technological engine, lagging behind LS only in the size of the community.
Found by chance and was fascinated. The engine that out of the box offers: a portal, personal and collective blogs, personal and collective photo albums, friends, clubs, ratings, guest profiles and file uploads in the same place, catalogs, FAQs, message boards, a store (modest in the kit and more serious in the form of separate paid component), a forum and a line with horns.The only thing missing is registration by invite. They correct me: there is already, albeit a third-party add-on, but very, very functional.
There are seventy tables in the database, but by load it is a perfectly sane solution, anyway for any social network you need a good server.
The biggest flaw: lack of unicode. The second - the table prefix is fixed. But the developers promise to fix it (they correct me, already there). The project is lively and the code is shameful. Those. I recommend it, especially if there are enough opportunities for the purposes, and there is not enough knowledge and time.
The organization of the admin panel and some logic are very similar to the jumla, sometimes it even seems that this is the assembly of the jumla with another admin skin and an already selected set of components, modules and plugins.
I don’t even know why I did not attribute it to multi-functional monsters.
Oh, I also found out what is written and will soon be released under this system a good forum. Although if you consider that even the built-in is quite good, then this one (with integration into the system) should generally be super.The forum is very modest, integration with a third-party is needed. In principle, the community has instructions for integrating with smf, but slightly outdated.
In general, the best BOX solution for a social portal.
The first two forums are paid. Voble (paid, alas) has plows of gallery, blogs, etc. IPB seems to be similar. Vanilla was originally written on webdvanol ideology. Very flexible architecture and plugin modifiability. I didn’t feel all three forums, but I plan to try vanilla. But in principle, a popular social network can be built on any good forum.
The most famous and common paid options. DLE is commonly used for warez and news underground portals. Plugins, skins and so on. I did not see either one or the other from the inside out, I mentioned it only because they are very common.
In the comments, I was provided with an addition on IPB and Voble . Thanks to LastDragon 'y and DevArt ' y.
0) Off. website: community.invisionpower.com/ (link to the forum itself)
1) It makes sense to buy only the third version (3.1.0 is coming soon)
2) Paid: IP.Gallery, IP.Blog, IP.Downloads, IP.Content
3) There are also free ones for clients: IP.Tracker (bug track, use it yourself), IP.Shoutbox (chat), etc. (can be found on their website)
4) It is very easy to develop modifications (minor changes to the functional; in most cases, you can do without modifying the forum code) and applications (such as IP.Gallery, IP.Blog, etc.). It is also very convenient to edit templates (after enabling development mode). Unfortunately, it requires a lot of manual work in the development and especially the assembly of releases, but most of me have already been automated.
5) They have all the documentation on the site (EN), can also be found in Russian (not all)
6) Disadvantages: there is copy-paste and govnokod (part - legacy), there are bugs, but they are quickly fixed. Poor BB-code parser (periodically updated).
7) Multilingualism (not complete - in ACP part of the lines can not be translated at once several languages, public part - all can be translated)
8) Skins support (+ mobile version, + xml version). IE6 is not supported. Out of the box there are bugs in IE7 / 8.
9) CNC (several types, the support depends on the specific application)
10) It is quite demanding on resources
11) Sphinx support out of the box (you can create plugins for your own applications)
12) Caching support - uses the database out of the box, but the necessary cache is included in one line (memcached, eaccelerator, etc.)
The official website here is vbulletin.org , the official developer community of vbulletin.com , Russian non-official technical support is vbsupport.org (the oldest and most comprehensive knowledge base in RuNet on this engine).
Initially, vbulletin was developed by the JelSoft company, later all products and the JelSoft company itself were purchased by InternetBrands. Accordingly, the developers have changed. By this time, the release of the vBulletin 4 line had ripened, and that did not bode well. As early as half a year, version 4 is still cheese, humiliatingly disgustingly compiled from what came out of the programmers. Even on the official community in the tag cloud, the vB 4 Is gay tag was fun for a long time. If you want to try the engine in business - I recommend version 3.8. Reliable, stable.
There are a lot of modifications for vBulletin. Some translated into Russian can be looked at for example here devilart.net/nashi-relizy-99 or at the same vbsupport.org, and some are created by Russian developers. Many paid and free hacks.
For vbulletin there is a fairly large number of skins, there are many large, well-known forums on this particular engine. VB has always been a leader in reliability and security in the field of forum engines. Relatively nice, clean code. Modifications, CNC are written simply. There are many components that make life easier for seachniks. The admin panel is simple and pleasant, multilingualism is supported. As a search, fulltext is used. There is a full Russian translation of the engine (admin / user part) for all versions.
Once upon a time the phrase an online store could only represent a subject. Pioneer or not, but this is the first open source store to gain popularity. Unfortunately, the release of osCommerce was already in 2003, and the next branch is still in rc status. Personally, I have no desire to poke around in the code that bears the legacy of those years of development. There it is difficult to suspect the existence of a good object model, separation into functional, display and data, and so on.
Since the license is gpl, over the years the store has appeared a lot of clones and forks, both free and very expensive. There are many communities, and in general - there are so many things around osCommerce that there is certainly an opportunity to find your personal Holy Grail. Only you need to look for it for a long time and hard.
Unlike osCommerce, Magento is a fairly young engine. At the Sourceforge Community Choice Awards 2008, Magento took first place in the "Best New Project" nomination. The engine is built on the Zend Framework, which immediately determines its monstrosity. He is big and heavy. But also powerful. But also raw. Those. the company that develops it takes money for customization and support, so there is no reason for them to make boxed candy.
The best choice if you know and be able to Zend Framework and not be afraid of incomplete Russification and gluttony. It makes sense on large projects.
This is a French shop, which personally would create enough problems for me. But the forces of the Russian community, the engine speaks Russian almost like a native.
It is put on Denver without any problems. The installer is good, with ajax checks. System requirements seem to be minimal.
From the first steps, a security concern is felt: the system forcibly requires renaming the admin directory and removing the install directory.
The possibilities at first glance are impressive.
More than 130 tables in the database. I’ll look at the load, but it moves quite quickly on the local machine. Although some complain of gluttony.
When unpacked, it occupies 14.5 MB, but more than three megabytes per tools, modules, img (demo data). One and a half meters js (jquery, tinymce and other stuff). 800 kb admin, 400 kb theme, 350 kb installer. 800 kb for 80 classes.
At first glance, pretty much OOP (php5), partly MVC, but not too ActiveRecord. I just don’t see that the model was separately issued.
Very serious service approach. The “Tools” menu alone contains: CMS for creating several static pages of the FAK type. There is a .htaccess generator and robots.txt. Database backup (directly in the store), work with subdomains, import from .csv, customizable quick access menu in the admin panel, customizable widgets, localization.
Naturally, a lot of settings. Localizers did a very good job, it’s only a pity that for Russia, not Ukraine. There are discounts, tax accounting, referral programs, etc.
Dofiga modules, four payment methods are supported (counting webmoney), statuses of goods (we are waiting for payment, the goods have ended, canceled, delivered, etc). Naturally, a bunch of statistics.
By God, at first glance I really like this store in terms of features, interface and service! Moreover, both for the administrator and for the buyer.
Chinese art. Immediately impressed: the first approach to the face gave 30 queries to the database, and the next - 4. It is also very economical with memory. Those. concern for performance, caching ...
The first call to the admin panel gave 28 requests.
Over 70 tables in the database.
And more people tested - recognized this store as the most economical for the server.
From the admin panel you can do optimization, backup and even direct queries: “SQL query works directly with the database. You must understand what you are doing. ”
There are crowns. There are, as in the presto, customizable quick menu items.
More advanced CMS: not just pages, but categories with a tree / type.
There is integration with forums, including wobble (3.x), ipb (2.1 \ 2) and phpbb (2.x).
Guestbook, user groups (customers).
Newsletters, affiliate programs, banners, SMS, delivery \ payment, bonuses, sales.
Translated not bad, but not as thoroughly as presto. However, work is ongoing. There is a nuance: translation 1.6.2, 1.7.0 and beyond - for a paid subscription. Therefore, I looked 1.6.1. Although later on nullaing I found a translation for 1.7.2.
If you choose between this store and presto, then I don’t even know what to choose. I liked Presto more, especially with the way it was translated and adapted, but here there is integration with forums and something else. In addition, strong caching is guaranteed here.
3.5 mb inclusions, incl. FCKeditor. Of these, 1.3 mb is something about the Chinese codepage. 2.8 mb admin. 1.5 mb theme. Some other data, js, api, wap, etc.
There is almost no OOP and the impression is that understanding architecture and writing modules \ etc will be much more difficult. But overall a worthy candidate.
The template engine seems to be smart, but something strange is there. In any case, the templates have a non-standard extension.
People who are horrified by osCommerce code and magento gluttony recommend OpenCart. The comparison table on the site promises that the engine can do more than osCommerce and prestashop. There is Russian, there are several dozen (maybe a couple of hundred in total) modules, templates, etc. The next time you search for the perfect store, this is a candidate for research.
There are plugins of different capabilities for different CMS. Often, if you already have a website and you need to add a store to it, it is best to find a plug-in for the CMS on which the website is built. For Drupal, these are: Ubercart (I recommend) and e-Commerce. For the jumla: virtuemart (there are no alternatives and I admit, the plugin is powerful, but more than a thousand files are something with something. Given several thousand files, jumlahs and the general slowness of the system, that of the plugin). There are a couple of plugins for WordPress. If you need a social network with a store - I recommend InstantCMS ...
The advantage is given to free engines. Paid will be considered only for comparison or from hopelessness, i.e. if there are no free analogues. Also a limitation on technology: php. About the engines on the pearl and python, I no more than heard, on sharpe and Java I was dealing with self-written ones.
So are being considered
- General purpose CMS. Information sites, business cards, blogs ...
- Multifunctional monsters.
- Galleries.
- Social networks.
- Forums
- Online stores.
General purpose CMS. Information sites, business cards, blogs ...
There are few comments, I am the moderator of the official forum and generally one of those who have been following this system from birth. Therefore, if anything, we will talk about MaxSite separately.
The system is written in a framework, therefore, you can add any functionality as you like. And the system itself on top of the framework provides a large number of service APIs. Therefore, there is nothing to be surprised that it will be found in other sections.
The structure and sharpening of the engine is initially a blog, but as an example of the same WordPress shows, this is not a limitation. Advantages: a good architecture and strong caching give good performance, ease of writing extensions - the whole system takes care of the service part, flexibility of settings - the conditions for displaying widgets, building links, constructing data types allow you to make sites very different in structure from blogs.
There are still some forks, someone else's heir, and so on. It seems that the neutrino is the heir to the mosquito, and the neutrino atomic editing is the fork of the neutrino. A CMS that does not use a database (and I needed it once). For simple business cards and blogs, it’ll work. Moreover, plugins can be written to it, and some of them are already available. There are publications feed, categories and tags, search (case-sensitive), templates, CNC. In atomics - in templates, you can control blocks (very convenient, actually).
Upd . Neutrino Atomic Edition will still be better, although even better see below.
I had to test the whole company: Neutrino Classic, Neutrino Atomic Edition, Nanote and Mosquito Blood Mary. Of all, their best impression with a significant margin was made by a mosquito. It all started with the installer - none of the above have it anymore. Further, absolutely all manipulations are done through the interface, and not manually. Everything is convenient and understandable. A minimalist file system, but for many cases it is enough. Plus the most friendly plugin creation documentation. And the possibilities are plus or minus all the same, maybe the mosquito is the most functional, and the nanote is the least functional. Blocks, tags, comments and everything about them, rss, cut, minimalistic file upload, etc. More about the engine .
One of the authors of this engine at one time created the Pikateka gallery, about which below. This is another engine that does not use a database. The engine on the files, but not the same as the above. As modules of this engine there is a forum (the most primitive), a catalog, ratings and even, it seems, a store. The engine also supports registration, user rights and all sorts of other things.
The problem is the same as with Picateca - the project died in 2007. Judging by the reviews, another problem, when used actively, indexes of the file database are damaged, i.e. Needs permanent backup. And I mention only then that sites that do not use databases are very well kept under version control systems.
The third Russian-language engine I met on Codeigniter. The first two are MaxSite and Cogear respectively. If MaxSite is basically blogging, a “wordpress killer”, and Kogir is a social network, then Image CMS, as you can see from a cursory glance, is more of a portal engine. At the moment, he is still sick with some childhood diseases, the first of which is few supplements. But otherwise it's a good engine, somewhat reminiscent of modx. The functionality is standard: category tree, tags, comments, widgets, photo insertion and gallery. Additional fields allow you to construct content types yourself. There is simple_cart. And the main advantage over the first two engines: multilingualism out of the box. True, I did not research it, but at first glance it was made better than in MaxSite.
- You can also use Wordpress, Drupal, Joomla, and almost any favorite CMS for general-purpose sites ...
Multifunctional monsters.
The versatility of monsters is that there are a lot of plugins for them, due to which you can build at least a portal, at least a multiblog, at least a store, at least a gallery, at least a social network, at least all of this at once.
Acquaintance began the year back in 2006 or 2007. Then the first branch lived and developed. The system was, as they say, not for the programmer. From the admin you can design absolutely everything. But at the same time, slowness was felt. One of the main drawbacks, but I still did not know that this drawback is the lack of Unicode. But at the same time there was a gallery (and not one), there was no way to organize CNC, multilingualism (buggy, but better than nothing), something else ...
Now I had to face again. Actual version 1.5.x. Unicode is. A bunch of already shitty plugins. CNC, forum (s?), Galleries, social network (s?), Store (s?) And so on. Slowness, in principle, remained, but still recommend the system to at least those who more or less know it and to whom it suits (or those who already have a working project on the jumble) - I agree. Surely caching can fix a lot.
As far as I remember, Russian-speaking communities competed among themselves. joom.ru, joomla.ru or something like that. Now there are considerably more of them. Which facilitates the search for information and answers to questions.
Upd. I had a chance to work with Jumla seriously. Impressions of the developer are absolutely obscene. The user experience is very favorable. You can design absolutely everything. But to choose from the same plugins the least load system, the least buggy and the one that will not be abandoned (or transferred to a paid basis) to the next version of the jumla ... Faster to write yourself (not for the jumble, of course). In general, if you have been familiar with dzhumla for at least a couple of years and know the picture and prospects of the plugins you need, then it’s good. And, yes, in addition you should have a powerful server that is not afraid to turn over CMS with several thousand files and monstrous requests such as fetching one record in a cycle.
Another problem is that a lot of the information that can be found relates to the jumel 1.0.x, while version 1.5.x bears the same name simply by mistake. Internally - this is a completely different system. And also presumably this year branch 1.6 should be released, which is already significantly different from 1.5.
Another update . I tried 1.6 beta 2. I twisted.
- What's good : more amenities in the admin area. Truly a much more powerful ACL. Explicitly for serious portals. It still needs to be studied and tuned. SEF out of the box.
- What is neutral : in terms of functionality, it’s the same jumla.
- What is promised : compatibility with extensions 1.5, tag hierarchy, many improvements in the framework, a new type of add-ons - libraries (used to be monolithic in the kernel), getting rid of the old heritage (and copy-paste) in the code.
- What's wrong : besides the ACL, a quick look did not find anything promised, a lot of bugs on the simplest things (the order of menu items does not change, query errors pop up on simple content hierarchies, etc.), there is no compatibility with components 1.5 at all (neither one has not even been installed).
- What's in retrospect : one major version in three years, a very loose community, many add-ons are abandoned in the version for j1.0, about which material is still appearing. The programming guide for j1.5 is still not fully translated into Russian.
- What's in the future : release 1.6 is unlikely to be released earlier than fall. Normally, they will run it in, port the main add-ons, write manuals and so on no earlier than a year later. Only then can serious projects be started on j1.6. Prior to this, only their own small sites, where everything is under control and in which case you yourself can fix the problem.
What I switched to, leaving Jumla - the wordpress. The admin of the WordPress after Jumla was the pinnacle of logic and simplicity. Moreover, then the branch about 2.0.x was relevant. Unicode was already supported then. Already immediately in the box were normal CNCs, customizable to the same. At the same time, CMS tossed and turned over FTP significantly faster than the dummy.
The more I learned about WordPress, the more I got into it. Plugin - a lot. This gallery and forum, and multilingualism, and all sorts of catalogs, and shops, and to hell with horns. You can do anything from WordPress. And with the WP MU assembly, you can also set multi-blog satellites.
But at the same time, as my knowledge improved and the versions of WordPress grew, I was more horrified by the growing gluttony. And limitations to some extent. On version 2.8.0, it was possible at certain times (when accessing the archive of a large site) to see 12,000 queries to the database. This is one of the most impressive, but not the only example of what happens when a project has no leader and the roadmap is determined by voting.
Again, this system can be recommended. But at the same time it is desirable to know and be able to tune to maximum performance. A big request to WordPress fans is not to arrange holivars here, because I express a subjective opinion, but confirmed by experience.
- Drupal .
Another monster that can do anything, which has tons of plugins for all occasions.
In the minuses, the great complexity of this system. Somehow it happened that for a long time trying to study it, I did not get along with it. The mice cried, pricked, but continued to eat the cactus.
And in the minuses there is a serious gluttony, not less, but sometimes even greater than the previous two engines.
The main advantage of Drupal is a very developed and well thought out hook system, which allows you to redefine almost any event in the system. Plus two of the most popular modules, cck and views, which allow you to construct arbitrary data types and their display.
Galleries.
Once I did not see normal galleries in other engines. Therefore, he studied individual galleries.
Either the first, or one of the first galleries that I met. There are some complaints about the interface, subjectively, I don’t like everything, but this is because the gallery is focused on multi-use, multi-album art and so on. User roles, user albums, limits, moderation and so on.
There are many translations, skins (tabular and artisanal), there are plugins. Integration with dummy and probably some other systems.
The branch I met with has been developing very sluggishly for several years, mainly bug and security fixes. The next branch has long been in svn, but only recently has grown to the status of beta and "official" download.
I would say that you can put it if the gallery is the main part of the site. Moreover, there is a plugin that implements the simplest cms, so as not to install any other engine for the sake of several pages. By the way, for the sake of several pages you can put neutrinos.
After at least four years of development, they posted version 1.5.x.
- Pikateka .
In contrast to coppermine, I really liked the picateca. Gallery is pretty simple. Easy fast and comfortable. Almost everything you need, almost nothing more. Instead of albums, tags. Multi-use is, but basic. The watermark in the code seems to be implemented, but it is not used anywhere by a separate function (by the class method), i.e. can not be enabled from the admin panel. And the main inconvenience: tag links use all the same tags, so for Russian ones links are terrible.
Abandoned at the end of 2006. No plugins are supported. The site has sunk into oblivion. If you merge from svn, then not the latest revision, but revision 179. By the way, there are bug fixes compared to the last archive.
I finished on her some little things. Porting a little bit to MaxSite. That just lacks a good gallery plugin.
Ps Now I found out that the author participated in the creation of ReloadCMS - a fairly large CMS that does not use a database.
- Gallery2 .
Monster from the gallery class. The code is ten times more than that of the galleries of all others. There is integration with different CMS. But since I never needed such power, I never really set it.
Now I looked, I saw that in February, after several years of development, 3.0RC1 was released. It turns out that the third branch was rewritten in kohan. As a result, she thoroughly lost weight. Another interesting thing: the project started ten years ago. Perhaps this is the oldest project php-gallery of the remaining afloat.
If you need a gallery without a database. Two small ones, one fork of the other. Abandoned in ± 2005. Simple, with basic capabilities (except that there were no lightboxes at that time, but it is not difficult to adapt it).
Simpleviewer is another thing without a database. Flash viewer, it was interesting just in the absence of lightboxes and other special effects. Although if you need a flash, you can adapt now. The viewer itself is free, but there is a paid version for it with companions, the lack of a link for downloading, etc., and a free plug-in for wordpress. I’m thinking about how to adapt it to MaxSite. But not very active, because for large volumes of photographs it is of little use.
Galleries without databases on the network can be found a great many, that sourssford, that phpclasses, that somewhere else. And to write or assemble such a script from pieces yourself is quite simple: navigation on the basis of existing folders, reading files in these folders and displaying thumbnails, a bootloader with preliminary hard-written authorization. And the minimum program is completed. You can attach a lightbox, and put watermarks during the loading process.
- Zenphoto .
Recently I came across such a swatch. There appear in plugins: cms, slideshow, googlemaps, flwplayer. We are talking about support for Ajax, multilingualism, skins, password protection access to albums. Presented awards that the gallery received. Development began in 2005 and continues to this day. There are rewards.
She didn’t want to bet on the third Denver with a five hundred mistake - she was missing something.
Social networks.
The functionality of social networks can be organized by any of the multi-functional monsters. And even some forums. I will especially mention MaxSite. While it has deployed one social network and is still doing. But the system is pretty not ready for this. You need to add a lot (architecture allows): admin panel for registered users (well, don’t let them in the admin panel of the site), ratings, downloads, so that everyone has isolated and no access to the main one, photo albums are also isolated from other people's interferences and tied to users, tree comments. Samborsky does not lay out his achievements, and to repeat them - there are more ready-made tools. Although I plan to still repeat my own development for and so that all the time I can use my favorite tool.
The first instance that I met in this field. Although, I lie, the first was drupal. :) So, in terms of functionality, this is quite an excellent social network. Almost full analog of a habr. Personal and collective blogs, ratings, topic polls, topic links, invites. Unfortunately, the gallery is paid, and even the topic-photo report is paid.
And so the system is an excellent platform for social networks and adding the necessary functionality. The most technological of those that I saw, and so on. There is actually a very good OOP that reminds me very much of Java. It can replace a forum if users are not too conservative to this and if you work a little with a file.
As for the components: Smarty is used for display, the client part is mootools. For working with DbSimple data, for caching DklabCache (memcache is highly desirable). To search for Sphinx, which accordingly must be installed on the server. Well, and a number of libraries.
Actively developing (from September 2008 to June 2010 940 commits, two active developers and several dozen third-party add-ons undergoing quality control). I recommend it if the functionality is enough for your purposes. The advantage over Drupal and the company is that Drupal needs to be added to the necessary functionality.
Now about the shortcomings. There are two of them. Despite the good optimization, for a full visited social network requires a dedicated server, i.e. system requirements are not the smallest (see paragraph about memekes and sphinx). And to refine the site on LiveStreet, you need to lay certain resources of time and money or independent efforts. This is not a wordpress for you, for which there are free plugins and templates, and for whom it’s not enough, there are ready-made assemblies for all occasions.
The engine, in spite of which the "Living Street" was created. Some time ago I was in limbo. Recently sold, after which its development resumed. But it seems to be agonizing again. The choice between the two is a matter of personal preference. But personally, I did not really look closely. Therefore, I only know that for a live street, blogs on subdomains are implemented by a paid hack, and for a big one, it seems to be standard.
Upd . The new developer evaluates the kernel as functional, but generally artisanal. He is thinking about correspondence of the kernel from scratch and possibly on the framework - Zend or Kohana. But the development speed still lags behind the development of LiveStreet.
- Explay .
Another instance. Each branch is rewritten again.
- iCore .
Fork Express. More corrected, more secure, and in the upcoming release pretty rewritten. But even though the author claims that the project is alive, evidence of this is not visible. The latest release and the last commit in svn in the fall of 2008. Recent publications on this subject in the spring of 2009. Further developments in the public are not laid out. The author’s sites on his own engine are buggy. The project is clearly dead, unlike its predecessor.
- Cogear .
Description rewritten . A social network written in Codeigniter. Of all the engines seen, the author of this has the most serious approach to documentation, appearance and all sorts of little things. In general, the author made a giant leap in writing this engine. A well-thought-out architecture allows you to do anything on Cogear if you have the knowledge, although the engine is geared to social networks. The disadvantage is that the entire author is drawn alone and obviously tired. Although it threatens to rewrite from scratch and not on CI. The plugin catalog now has a dozen add-ons and not a single topic, but writing extensions is very simple and fast due to the architecture and the hook system. Also, jambs sometimes happen, and the engine is completely not friendly with wine hosting. But the author reacts very quickly to bug reports.
Along with LiveStreet, it is a very technological engine, lagging behind LS only in the size of the community.
Found by chance and was fascinated. The engine that out of the box offers: a portal, personal and collective blogs, personal and collective photo albums, friends, clubs, ratings, guest profiles and file uploads in the same place, catalogs, FAQs, message boards, a store (modest in the kit and more serious in the form of separate paid component), a forum and a line with horns.
There are seventy tables in the database, but by load it is a perfectly sane solution, anyway for any social network you need a good server.
The biggest flaw: lack of unicode. The second - the table prefix is fixed. But the developers promise to fix it (they correct me, already there). The project is lively and the code is shameful. Those. I recommend it, especially if there are enough opportunities for the purposes, and there is not enough knowledge and time.
The organization of the admin panel and some logic are very similar to the jumla, sometimes it even seems that this is the assembly of the jumla with another admin skin and an already selected set of components, modules and plugins.
I don’t even know why I did not attribute it to multi-functional monsters.
In general, the best BOX solution for a social portal.
- vBulletin, IPB, Vanilla .
The first two forums are paid. Voble (paid, alas) has plows of gallery, blogs, etc. IPB seems to be similar. Vanilla was originally written on webdvanol ideology. Very flexible architecture and plugin modifiability. I didn’t feel all three forums, but I plan to try vanilla. But in principle, a popular social network can be built on any good forum.
- Social Engine . DataLife Engine (DLE) .
The most famous and common paid options. DLE is commonly used for warez and news underground portals. Plugins, skins and so on. I did not see either one or the other from the inside out, I mentioned it only because they are very common.
Forums
- For consideration: phpbb, other bb, vanilla, smf, something else. PhpBB - very heavy, sophisticated, powerful. Very difficult to administer (I even dreamed in nightmares). For simple cases I do not recommend.
- In contrast, punbb is lightweight, nimble and simple. It’s a little unusual that the extensions should be in xml files, but it turned out pretty quickly. In miniature and simplicity lies two tricks. First: if there is no functionality, then it is not. Second: many extensions modify the database structure and commit other bullying on the basic delivery of the forum. Forum - for simple cases - that's it. And judging by forum.mozilla-russia.org - you can take a chance and put this forum for a serious project. Just keep in mind that forum.mozilla-russia.org has its own assembly of this forum, which, it seems, is not compatible with the original. But open.
- Vanilla (I have not tried it, but I will definitely try it) is a forum that was initially declared "not like the others." Due to the modularity and more than 450 plugins, you can create forums of very different functionality and focus. The declared ease of the forum itself and ease of study, originally webdvanol ideology, its own framework for writing extensions, the ability to embed the forum in a variety of engines. Version 1.1.10 was released in November 2009. Version 2.0 is available in the repository, they say, RC has just appeared. The Russian community is modest and half dead, but they have hopes for a deuce.
- Forum plugins are in wordpress, drupal (in the standard package), it seems, in jumla ...
- For MaxSite, a primitive and very buggy forum in a limited circle is another one being written. Perhaps it will develop, including my efforts. Up to functionality roughly punbb. But it will not be soon, if at all.
In the comments, I was provided with an addition on IPB and Voble . Thanks to LastDragon 'y and DevArt ' y.
0) Off. website: community.invisionpower.com/ (link to the forum itself)
1) It makes sense to buy only the third version (3.1.0 is coming soon)
2) Paid: IP.Gallery, IP.Blog, IP.Downloads, IP.Content
3) There are also free ones for clients: IP.Tracker (bug track, use it yourself), IP.Shoutbox (chat), etc. (can be found on their website)
4) It is very easy to develop modifications (minor changes to the functional; in most cases, you can do without modifying the forum code) and applications (such as IP.Gallery, IP.Blog, etc.). It is also very convenient to edit templates (after enabling development mode). Unfortunately, it requires a lot of manual work in the development and especially the assembly of releases, but most of me have already been automated.
5) They have all the documentation on the site (EN), can also be found in Russian (not all)
6) Disadvantages: there is copy-paste and govnokod (part - legacy), there are bugs, but they are quickly fixed. Poor BB-code parser (periodically updated).
7) Multilingualism (not complete - in ACP part of the lines can not be translated at once several languages, public part - all can be translated)
8) Skins support (+ mobile version, + xml version). IE6 is not supported. Out of the box there are bugs in IE7 / 8.
9) CNC (several types, the support depends on the specific application)
10) It is quite demanding on resources
11) Sphinx support out of the box (you can create plugins for your own applications)
12) Caching support - uses the database out of the box, but the necessary cache is included in one line (memcached, eaccelerator, etc.)
The official website here is vbulletin.org , the official developer community of vbulletin.com , Russian non-official technical support is vbsupport.org (the oldest and most comprehensive knowledge base in RuNet on this engine).
Initially, vbulletin was developed by the JelSoft company, later all products and the JelSoft company itself were purchased by InternetBrands. Accordingly, the developers have changed. By this time, the release of the vBulletin 4 line had ripened, and that did not bode well. As early as half a year, version 4 is still cheese, humiliatingly disgustingly compiled from what came out of the programmers. Even on the official community in the tag cloud, the vB 4 Is gay tag was fun for a long time. If you want to try the engine in business - I recommend version 3.8. Reliable, stable.
There are a lot of modifications for vBulletin. Some translated into Russian can be looked at for example here devilart.net/nashi-relizy-99 or at the same vbsupport.org, and some are created by Russian developers. Many paid and free hacks.
For vbulletin there is a fairly large number of skins, there are many large, well-known forums on this particular engine. VB has always been a leader in reliability and security in the field of forum engines. Relatively nice, clean code. Modifications, CNC are written simply. There are many components that make life easier for seachniks. The admin panel is simple and pleasant, multilingualism is supported. As a search, fulltext is used. There is a full Russian translation of the engine (admin / user part) for all versions.
Online stores.
- Several for consideration, but meanwhile the same MaxSite.
The store for him is written by mein a preliminary build posted on the off-forum (my version most likely will not see the light of day). Small and with basic functionality. You can use it only if you need a small store and it is on the max website. In other cases, I recommend prestashop or ecshop. Or magento.There are chances for the appearance of another plug-in store - mentions flickered(it was he who had already appeared). - osCommerce and its clones and forks.
Once upon a time the phrase an online store could only represent a subject. Pioneer or not, but this is the first open source store to gain popularity. Unfortunately, the release of osCommerce was already in 2003, and the next branch is still in rc status. Personally, I have no desire to poke around in the code that bears the legacy of those years of development. There it is difficult to suspect the existence of a good object model, separation into functional, display and data, and so on.
Since the license is gpl, over the years the store has appeared a lot of clones and forks, both free and very expensive. There are many communities, and in general - there are so many things around osCommerce that there is certainly an opportunity to find your personal Holy Grail. Only you need to look for it for a long time and hard.
- Magento
Unlike osCommerce, Magento is a fairly young engine. At the Sourceforge Community Choice Awards 2008, Magento took first place in the "Best New Project" nomination. The engine is built on the Zend Framework, which immediately determines its monstrosity. He is big and heavy. But also powerful. But also raw. Those. the company that develops it takes money for customization and support, so there is no reason for them to make boxed candy.
The best choice if you know and be able to Zend Framework and not be afraid of incomplete Russification and gluttony. It makes sense on large projects.
- Prestashop . A project that I really liked.
This is a French shop, which personally would create enough problems for me. But the forces of the Russian community, the engine speaks Russian almost like a native.
It is put on Denver without any problems. The installer is good, with ajax checks. System requirements seem to be minimal.
From the first steps, a security concern is felt: the system forcibly requires renaming the admin directory and removing the install directory.
The possibilities at first glance are impressive.
More than 130 tables in the database. I’ll look at the load, but it moves quite quickly on the local machine. Although some complain of gluttony.
When unpacked, it occupies 14.5 MB, but more than three megabytes per tools, modules, img (demo data). One and a half meters js (jquery, tinymce and other stuff). 800 kb admin, 400 kb theme, 350 kb installer. 800 kb for 80 classes.
At first glance, pretty much OOP (php5), partly MVC, but not too ActiveRecord. I just don’t see that the model was separately issued.
Very serious service approach. The “Tools” menu alone contains: CMS for creating several static pages of the FAK type. There is a .htaccess generator and robots.txt. Database backup (directly in the store), work with subdomains, import from .csv, customizable quick access menu in the admin panel, customizable widgets, localization.
Naturally, a lot of settings. Localizers did a very good job, it’s only a pity that for Russia, not Ukraine. There are discounts, tax accounting, referral programs, etc.
Dofiga modules, four payment methods are supported (counting webmoney), statuses of goods (we are waiting for payment, the goods have ended, canceled, delivered, etc). Naturally, a bunch of statistics.
By God, at first glance I really like this store in terms of features, interface and service! Moreover, both for the administrator and for the buyer.
- ECShop .
Chinese art. Immediately impressed: the first approach to the face gave 30 queries to the database, and the next - 4. It is also very economical with memory. Those. concern for performance, caching ...
The first call to the admin panel gave 28 requests.
Over 70 tables in the database.
And more people tested - recognized this store as the most economical for the server.
From the admin panel you can do optimization, backup and even direct queries: “SQL query works directly with the database. You must understand what you are doing. ”
There are crowns. There are, as in the presto, customizable quick menu items.
More advanced CMS: not just pages, but categories with a tree / type.
There is integration with forums, including wobble (3.x), ipb (2.1 \ 2) and phpbb (2.x).
Guestbook, user groups (customers).
Newsletters, affiliate programs, banners, SMS, delivery \ payment, bonuses, sales.
Translated not bad, but not as thoroughly as presto. However, work is ongoing. There is a nuance: translation 1.6.2, 1.7.0 and beyond - for a paid subscription. Therefore, I looked 1.6.1. Although later on nullaing I found a translation for 1.7.2.
If you choose between this store and presto, then I don’t even know what to choose. I liked Presto more, especially with the way it was translated and adapted, but here there is integration with forums and something else. In addition, strong caching is guaranteed here.
3.5 mb inclusions, incl. FCKeditor. Of these, 1.3 mb is something about the Chinese codepage. 2.8 mb admin. 1.5 mb theme. Some other data, js, api, wap, etc.
There is almost no OOP and the impression is that understanding architecture and writing modules \ etc will be much more difficult. But overall a worthy candidate.
The template engine seems to be smart, but something strange is there. In any case, the templates have a non-standard extension.
- OpenCart .
People who are horrified by osCommerce code and magento gluttony recommend OpenCart. The comparison table on the site promises that the engine can do more than osCommerce and prestashop. There is Russian, there are several dozen (maybe a couple of hundred in total) modules, templates, etc. The next time you search for the perfect store, this is a candidate for research.
- Plugins
There are plugins of different capabilities for different CMS. Often, if you already have a website and you need to add a store to it, it is best to find a plug-in for the CMS on which the website is built. For Drupal, these are: Ubercart (I recommend) and e-Commerce. For the jumla: virtuemart (there are no alternatives and I admit, the plugin is powerful, but more than a thousand files are something with something. Given several thousand files, jumlahs and the general slowness of the system, that of the plugin). There are a couple of plugins for WordPress. If you need a social network with a store - I recommend InstantCMS ...