2014 Will Not Be the Year of Flat File Websites
I’ve seen a few people proclaiming that 2014 will be the year that flat file websites become popular.
After testing a good number of flat file website tools, I completely disagree. These tools are unlikely to become popular because they are being really badly marketed. Flat file sites are currently only selling themselves to a niche groups of developers, rather than the mass market.
Let’s take a tour of some of the best flat file website creators:
Getkirby.com has the slogan, “Kirby is a file-based cms” and secondary tag-line, “No database, just files and folders”.
Statamic.com says, “Pack up your content, markup, and style. Leave the database behind.” Then it says “Making websites just got fun again” and says it was built by and for developers and designers.
Pico.dev7studios.com say that Pico is a “stupidly simple & blazing fast, flat file CMS.” It then starts talking about no database, markdown and Twig.
Octobercms.com says that it’s “A minimalistic platform that gets back to basics.” October says it “can be understood at a glance” and then starts talks about “Page Components” and an “Extensible platform”.
Jekyllrb.com says “Transform your plain text into static websites and blogs.” It says “Get up and running in seconds” and then shows the command line.
Do you notice who the target audience is for all these platforms? Developers.
None of these platforms are talking to ordinary users:
- Users don’t care if the database is in a file or in MySQL.
- Users don’t care about “big” or “small” systems.
- Users don’t care about markdown. They’ve never used it before and will almost certainly perfer a WYSIWYG editor.
All of these platforms are by developers and for developers. None of them are talking to a wider audience.
Are you saying these are bad platforms?
No. Some of them are really good platforms.
What I’m saying is that they are absolutely not on the path to popularity.
These are great platforms, but only for a niche audience of developers and designers.
Will static websites ever become popular?
Possibly, in the future. I’d love to see some successful static, popular website tools.
However, in addition to solving the wrong problems, statics website generators are held back by several other problems:
- Many have commercial cores, which hampers adoption. Why not give away the core and sell add-ons?
- There are too many options. At some point one or two of these generators may break from the pack, as WordPress, Drupal and Joomla did with dynamic CMS’s. But it hasn’t happened yet.
- No community. There are no events, no marketplaces and little in the way of discussion forums for many of these products.
- Poor or non-existent documentation.
So who does get it right?
It’s worth comparing the generators above to Ghost.
Ghost.org proclaims that it is, “Just a Blogging Platform” and has these delightfully simple instructions: “Just want to get a blog up & running? Hit the green button. Have your own server & want to install Ghost on it? Hit the blue button.”
Ghost has a marketplace already Marketplace.ghost.org and is trying to build up a community. They have a forum at Ghost.org/forum and now have over 100 themes on ThemeForest. Whether you like Themeforest-style sites or not, getting featured on those sites is an indicator of popularity and support. Also, rather than try and sell the core files, they made money first via fund-raising and now via a hosted service.
Ghost emerged from the WordPress world and you can tell they’ve learned many lessons from WordPress.
It’s very early to expect widespread popularity for Ghost in 2014. Ghost is still at version 0.3, it still used markdown rather than a WYSIWYG and it does use Node.js for which hosting options are still limited.
But, looking ahead, if any of these static site creators has the chance to become used by anyone except developers, it’s probably Ghost.
Hello – I’m from Perch and we’re not a static CMS. Perch requires PHP and MySQL. We have post on our site explaining why we think a database is important [url=http://grabaperch.com/blog/archive/why-does-perch-use-a-database]http://grabaperch.com/blog/…[/url] and our lead developer has explained the technical issues in more depth here [url=http://allinthehead.com/retro/362/why-use-a-database]http://allinthehead.com/ret…[/url]
We do aim our product at designers and developers, but by creating a great content editing experience so they can focus more on creating websites, not on training clients. Much of the time the designer makes the choice of CMS, and many studios build their offering around a particular CMS or group of products as everything – however small – has some learning curve involved.
Jekyll is the only one in this list that creates static sites.
Admittedly, Perch does render *some* html in advance and cache it, but it’s still generating content on request so a long way from being a static site generator. The others in this list are as dynamic as Joomla, WordPress et al, just using a different (dynamic) method to render an output.
Thanks, Seth. You’re right – I tweaked the phrasing and a couple of the examples. When I started writing my examples were mostly static sites, but by the time I finished they were mostly flat file sites.
To be honest, the static sites were were even worse examples of the thesis above. I left many of the list because they were no more than a Github profile.
Thanks Steve.
FWIW, I think it depends on the user and use-case. For a developer, flat file makes a lot of sense. I can version control *everything* for starters. Though as soon as content volume increases it’s hard to get away from the fact that databases are incredibly efficient way to store and retrieve data.
Yes, you’re right there, Seth. There’s a lot of compelling reasons for flat-file systems.
It’s just that most (all?) of them benefit the devs, not the users.
Backing up your site or moving hosts is certainly much easier when it’s a simple copy/paste operation. I’d say that security, of the site and server, also benefits from not having a database; it’s often targeted as an entry point.
Jekyll excels in that respect, as it can run on a server which doesn’t have either PHP or MySQL available. That doesn’t leave much in the way of vulnerabilities to exploit.
You raise an interesting point about usability and targeting for non-technical types. Keep an eye on [url=http://OctoberCMS.com]OctoberCMS.com[/url] as it is only in the embryonic stage of development and it hopes to solve a lot of the problems you are describing. Albeit targeted towards developers initially, these developers will eventually build the tools to make the platform accessible to “anyone except developers”.
This is the first I have heard of “flat file”. Is the definition of “flat file” a site that is not database driven? Or is there more to it than that? Sorry to be so ignorant but I won’t be after this! (:
Hey Toni. Yes, that’s exactly it. Imagine an updated version of Dreamweaver.
I guess the fact you, as a web savvy person, hasn’t heard about the platforms, is another reason they won’t be popular soon.
Thanks Steve for the explanation and the compliment. I just added your blog to my feeds. I appreciate all your great content it helps to keep me up to date. Happy New Year.
Thanks Toni! I hope to see at an event again soon.
I haven’t seen any of these proclamations about flat file systems becoming popular with the masses. It surprises me that anyone would think that. My take is they might become popular with elite niches because of the masses. I think that is their actual goal. WordPress, Joomla and Drupal have a high cost of ownership over time, especially the last two due to their “breaks everything” upgrades every few years. They all have inherent performance and security issues that can simply be bypassed with a flat file system. Some of this has to do with designers who aren’t developers or who have limited programming skills wanting to serve clients without having to bring all those skills and hats together.
You seem to have overlooked one key difference about a few of these systems where they were developed to handle structured content in a very forward-thinking way that just cannot be done well or easily, certainly not by default with the most popular CMSes. There may also be an interesting split afoot where the older CMSes turn toward becoming application platforms and leaner, often flat file systems cater to designers and users who just want a simple, solid content site without a sea of javascript and application code to maintain.
I’ve been curious too about the proliferation of startups offering cheap flat file hosting via dropbox and/or AWS which is all about catering to designers handcoding sites, or using Jekyll or a really simple CMS and templating system the host provides. These are niche markets, but they may be substantial in value/profit and not all that small. Time will tell.
Thanks Dan
Yes, some of those Dropbox-based CMSs are fascinating, eliminating even the need for FTP.
A couple of examples:
[url=http://droppages.com/]http://droppages.com/[/url]
[url=http://brace.io/]http://brace.io/[/url]
imagine being able to do everything from a Dropbox file on your desktop. No FTP, no databases, no migrations. I think those are the kinds of benefits that an ordinary user would love.
My favorite database-less CMS was simple PHP blog, see: [url=http://sourceforge.net/projects/sphpblog/]http://sourceforge.net/proj…[/url] It’s been around for years and is dead-simple and rock-solid. But really, how many of these type products does the universe need?
Not many, but there are 1000’s of them!
You nailed it >>These tools are unlikely to become popular because they’re solving problems that very few people care about.
I think they have their time and place: for example I’m looking to build up some system reference document sites with a flat file system because I don’t need the whole shebang.
It would be interesting to see a primarily flat file design that is built to utilize databases where beneficial. You always see one vs the other but I am sure there could be some sort of middle ground.
DokuWIki is a wiki based on flat files which can be used as a blog too with plugins. I for one being a developer just love flat file CMS. But you are probably right about the average user.
Hi there. I think it has to do on which market you serve. Yes, at some point hosting companies were offering unlimited sql services. But the last couple of years in my experience I ran into hosting companies more than once that did not support CMS any longer on that particular hosting deal (even though they offer SQL along with it) Their answer: You should switch to a different more optimized hosting package (read more expensive). People who run small businesses do not get that at all. And to be hones me neither. These people just want to update their website in and easy way. And if a flat file CMS makes this possible (because they don’t need all the fancy plugins and extensions) than thats ok. So yeah, I honestly do believe there is a market for flat file CMS. Some CMS’s became too bloated and some hosting companies smell money out of it. Besides that you can cut development cost and offer a cheaper deal. And that counts as well I think 😉
Hi Biana
Yes, we’re on the same page, I think.
There is a market for flat file CMSs and they have advantages for end-users.
My thinking is just that existing flat file CMSs are not positioning themselves well to appeal to those end-users.
I agree that the flat file CMS devs have a (too) strong technical focus so yes were on the same page. But as a communication professional I have no trouble in making the translation to the end user for them. 50% of my clients run their website on flat file CMS now. So yeah maybe the flat file CMS won’t make in 2014 but it sure will be in 2015 …..if I work hard enough ;-). Now kidding aside, I expect this will change soon enough.
Spot on!
I guess this fad is all about a rearguard action from some nostalgic developers.
In any case a few of these flat file websites advocates have been drinking some strange brew. Read this, found on [url=http://appstorm.net]appstorm.net[/url] about WordPress in an article about the superior values of Kirby. A master piece of bad faith.
“Digging into its code seems almost as daunting as patching the Linux kernel”
“Want to manage your images and uploads? Again, it’s WordPress’ interface, or digging into the file structure WordPress setup”
“if you need to edit your articles, expect to either have to dig into MySQL if you don’t want to use WordPress’ built-in editor”
OMG there are several tens of millions of MySQL experts using WordPress and we did not know it ! What is this nonsense?!!
Almost fell of my chair, laughing out loud.
The author of these daring statements calls himself a developer and a long time WP user… no comment.
Can’t resist: here’s the url:
[url=http://web.appstorm.net/reviews/web-dev/kirby-the-flat-file-cms-thatll-make-you-rethink-content-management/]http://web.appstorm.net/rev…[/url]
Enjoy!
You completely miss the mark; websites for the public are built by developers. So if developers start moving towards flat systems, so will ‘the public’.
Agreed; Steve even says it at the start of his run down:
“Users don’t care if the database is in a file or in MySQL.”
But developers do! We care how we get our job done. With small business websites, it’s always the same story: person contacts developer asking for a website, says they want it to be editable so they can change things. How that editable feature is achieved is open to debate and often the client is asking the developer for advice.
So yes, exactly, the users/clients don’t care how a website is made editable so long as it’s easy and they don’t feel overwhelmed. 🙂
Great post – I do see some truth in it but still hold flat file cms have great benefits I have one test blog in Phile cms and it is great. The main benefits are:
– Speed, google loves this and this will be big in the future as we know mobile is the platform everyone is looking to.
– Security as there is no connection to exploit it is secure something which you have to be on top of if you are on wordpress.
– Cost if someone needs a simple small site I think these are the real winner, okay yes they need to come a bit further but you could get a good bespoke fast site from these without the need for in depth and costly themers and or developers to get your site up in drupal or wordpress
Overall it has its place and I just hope its market expands and loses the bloat that cms’ have these days, clients should spend the money on better designs and products not maintenance of big cms’ for small sites.
Steve
Thanks Steven. I’ll give Phile a test-drive.
There’s a good chance that 2015 or 2016 maybe a growth year for these platforms. But to do so they’ll need to get a lot better at explaining themselves to more than just devs.
I am personally evaluating a move to one of the flatfile cms’s or at least a file based database CMS. The reasoning for this is simple…bloat.
I started life on e107, moved to Joomla for several years and in 2012 moved to wordpress. The reasoning for this, is that as my site matured, I realised that whilst I loved the content of the site, my audience preferred to spend their time in the forums. Whilst they still use the frontend, the reality is that they don’t need the level of interaction and bloat that WordPress offers.
Lately I ran into a resource issue on my site, memory pushing out to 100% consumption for extended periods on a 4GB RAM dedicated server. That just wasn’t acceptable. I ended up shutting down WordPress to see if this was the cause, and my memory eventually dropped down to 40% once the queue cleared. I enabled it again 2 hours later only to see the same consumption.
Whilst some of this was plugins, even after disabling all the plugins and reverting to original themes I saw ridiculous amounts of memory being consumed.
The net result being that this made me look at just what my front-end does, and the database purely exists as a repository for the data which doesn’t really need to be sent dynamically.
I think that if people really dive down to the nitty gritty of what their site does, the larger CMS’s are starting to miss the mark and move to a be all and all when their target audiences don’t really need that.
Those projects have become behemoth products as ideas push beyond their niche.
When WP started it was a simple blogging software, it’s now grown to a full blown CMS, people have stayed on it, “just because” but if you do a real needs assessment, does it really still suit the need.
Actually the tool that I am using right now, Disqus, is an example of where the likes of WP are no longer the right tool for the job. Resource wise, a simple flatfile system or file based database with lower memory imprint, would be more cost effective and faster for you, you could have an internal comment system but if you are using disqus, why bother?
I could go on, in more detail, whilst from your perspective they are filling a requirement that people don’t need. The reality is that they are filling a very real need, and this is why they are rising in usage. They may be aimed at developers, but as long as the interface is simple, the end user doesn’t care what system is there, and for developers it means less server pains.
I will say however some of them are bizarre in their methodologies and rigid enforcement of “ideas” for instance forkcms telling you that you need to install in the root because they don’t agree otherwise, which is ridiculous, I have sites where blogs are a part of the site, but not part of the main site, internal company sites are another example and if they want to move into say the education space with LTI is another,
anyway, I don’t know if 2014 is the year, but it’s certainly the year I am strongly evaluating them for the content aspect and front-end of my site and leave the more dynamic aspects to my forums.
Thanks Dan, what a great, interesting comment.
I don’t disagree. A new generation of simpler systems will probably take over. Personally, I don’t think they’re quite ready yet, but the opportunity is there.
hmm maybe I need to start development on a more mature targeted version.
The big question for me is theming here and that is where my time will be spent. I push and pull data between the forums and wordpress, I also change my theme every year or so to keep the site fresh for the audience that demands it.
I want to make sure that the theme is tackled in a dynamic fashion so i do not need to make wide-scale changes with design change.
Im interested in seeing how data is shared dynamically as well, this is why the XML based sites seem better suited as I can use these as a basis for grabbing the content “remotely”
I find “flat-file” CMS’es super useful for small- to medium websites. They often cut a lot of the fat away, and are easy to work with for basic implementations, leaving the customized output fast and effective. I am developer/designer, but consider myself more a designer, and many of my designer colleagues enjoy working with flat-file CMS’es.
Flat-file CMS’es have several limitations, but in an age where the trend is to make websites clean and simple (across all devices), this fits the bill very well for a large niche of small- to medium websites. Besides, they are faster than DB websites which is just another layer of data to load often from another server … I have had my share of DB issues and connection issues with DB-powered websites (specifically WP, on a shared hosting environment @MT).
Well, Monstra is incredibly simple and is specifically targeting users. It feels to me like WordPress does but with only absolutely bare essential features. Just throwing it out there that not all FFCMS’s are walking the dev only path…
I agree with what you say to a point. But I can see File based CMS systems taking off simply for the fact many people want a lightweight solution to making sites and don’t want to mess around with databases. There are plenty of security risks with wordpress if you don’t keep up to date. Not having a database could make your site a bit more secure then it would be with a database. Also, typically a simple CMS like a file based one is simpler to design for unlike wordpress, where you need to have a good understanding of it to write your own themes. Many people don’t know how to install wordpress, so developers typically install it for them, same would be the case with something like this.
I am writing my own file based CMS simply for the fact I can. Doing such a project even though they already exist helps me learn the language i am writing it in.
Speed and simplicity are two factors that are overlooked. Not every lightweight 5 page website needs WP / Joomla / Drupal – it’s a massive bloated overkill.
+1, and I’d add security to that list too – there’s not much to ‘hack’ when a site is effectively just HTML.
Not sure if I think this article is interesting or backwards. I agree the existing flat CMSs aren’t there yet, but don’t judge the concept just on the current attempts. Of course they are aimed at developers, they are the early adopters of new web technologies, and generally the people who develop websites and will likely propagate flat CMSs to the general consumer market.
I agree with most of your points though, that’s why I actually started developing my own flat CMS (//[url=http://flot.io]flot.io[/url]), it’s not ready yet, but will be targeting the ‘everyday user’ and making a ‘system’ anybody can hopefully use!
Ghost has it right? Really? I was one of the kickstarter backers with ghost. If there is anything on this page aimed at Devlopers it’s Ghost. As primarily a front end html/css developer / designer I thought it’s be awesome. Turns out the thing needs a bunch of specialized crap to install and can’t run on most servers. Yes, it looks fantastic but Kirby & Statamic I can run nearly anywhere. Even trying to install custom themes on Ghost’s servers is a hassle. Ghost, upon release, made me feel duped that I backed it.
I have to agree with this sentiment. It’s even worse when you try to run more than one site, or host Ghost along with WordPress installs. On nginx. Ugh.
Until they got their own hosting plan up and running I would never have recommended Ghost for non-developers. Oh, and also — where is the fancy dashboard they touted so highly early on? Oh, I see. Pushed back to 2015 now? Version 1. They are currently on 0.5 so…
I don’t think statamic is aimed at developers. It’s got a great admin interface (even a wysiwyg if you must) for content editors & the template system is super easy for designers to make exactly what they want & not have to get involved in databases & php. The fact its got a great API for plugin development is a bonus for when you do want to involve a developer.
Yes, but fundamentally users don’t get to choose whether their site ought be built on WordPress or Statamic. Ultimately it’s the developer’s/designer’s choice.
Flat file web sites are a branch of flat trend. Flat trend is about tackling bloat-ware that plagued website development for many years. The bloat became apparent with the rise of mobile devices and consciences toward bandwidth usage. Now is the time when designers and developers are realizing we need to lean down; we need to get back to building performance, not feature-rich, websites.
Just looking around, I stumbled across Pico CMS, a “stupidly simple” CMS that has me learning the Twig templating before I can use it. It says, ” there is no administration backend . . . You simply create .md files in the “content” folder and that becomes a page.” Why don’t I simply create .htm files in the public_html folder and let those become pages?
I still haven’t found anything that hits the sweet spot better than Wolf CMS. It doesn’t use any templating language, but it does provide a few constants and functions, just enough to navigate and manipulate pages. End users entering content can use the smart editor it includes, so they don’t need to know any HTML, though knowing a few HTML tags is hardly a huge burden. End users certainly don’t need to know PHP.
I created a ‘flat file cms’ because I’m a developer and it was a nice academic challenge. You can read about it here: [url=http://www.kestrelblackmore.com/projects/blogmatrix]http://www.kestrelblackmore…[/url].
It does have some advantages:
– It produces static html. Can’t get any faster than that.
– Because it’s static html I can host it for free with GitHub Pages ([url=http://pages.github.com]http://pages.github.com[/url]).
– It uses git for deployment.
Having said all that it’s definitely not for mass consumption 🙂
Good sense of clarity brought to a subject spurred by poorly devised shock and awe headline writers seeking traffic to waste our time. Ghost looks promising, but IMO it’s under delivering on it’s promises. It was supposed to have a dashboard months ago, still doesn’t have the dashboard that they feature on their site. Ironically, I feel it’s ghost’s transparency that’s lacking. I was mislead when I first started using it. It’s nowhere near the level it purports. +Green
And what about Monstra CMS? 😉
It’s user friendly, just download -> unpack, upload on server, name your page and you’re good to go.
[url=http://monstra.org/]http://monstra.org/[/url]
Pulse is my favorite flat file cms for small to medium sites. Very simple to use with or without their template and they use Imperavi’s Redactor as the editor. Our clients love the user friendly backend – [url=http://pulsecms.com]http://pulsecms.com[/url]
Have you tried [url=http://razorcms.co.uk]http://razorcms.co.uk[/url] a file based CMS that tries to talk more to the average user. Not built for programmers, but able to be altered by programmers. They even offer auto installs when you get the hosting account through the website, all setup without any fuss or anything to do on your part.
getsimple is the one that stands out for the moment, razorCMS might become a challenger but right now lacks too much functionalities.
Bad news. Ghost has gone terminal 🙁
Seriously, when will these platforms get it that no one besides developers want to mess around with Node.js and/or Terminal crap? The Terminal is the exact reason I switched back to Windows from Linux.
I hope someone can create a true “flat file” platform.
good article. thanks.
I have spent 2014 thus far developing in WordPress but am gearing up to do my own site. After some reading and research, before landing here, I have decided to try Statamic to build my own site. Aside from flat files, the version control is very appealing.
This article is on point: the marketing is not geared towards the masses of end users. This approach (and the associated marketing) to CMS is definitely geared towards the developer/designer: myself more of the latter.
For those who already get it (that database cms systems are overkill for small to medium sites without super-complicated needs):
Premium:
– Kirby ([url=http://getkirby.com/)]http://getkirby.com/)[/url]
– Statamic ([url=http://www.statamic.com/)]http://www.statamic.com/)[/url]
– many others, these seem to be the top two
Open Source:
– Grav ([url=http://getgrav.org/)]http://getgrav.org/)[/url]
– there are others, Grav seems to have the best documentation and will soon have a great admin panel (preview here: [url=http://getgrav.org/blog/admin-plugin-development)]http://getgrav.org/blog/adm…[/url].
note: I’m in no way connected to any of these.
It has been about a year now and Grav has really grown from a CLI developer’s tool to a flat file CMS any user could use. I like the fact the team is looking into incorporating their PHP framework, Gantry into Grav making the admin and thus usability by users much greater and easier. Grav and Gantry are both the work of Andy Miller and volunteer work by most of the team at RocketTheme: [url=http://www.rockettheme.com/the-team/]http://www.rockettheme.com/…[/url]
[url=http://www.getgrav.org/]http://www.getgrav.org/[/url]
[url=http://www.gantry.org/]http://www.gantry.org/[/url]
However, while I agree with the author’s stance on who these software applications are developed for let us not neglect the ever increasing harm of users getting a “plug and play” CMS like WordPress, doing some work and then expecting magic. These are tools and as such it is a developer’s prerogative to ensure the user understand such. As a tool, they require the proper uses to make them work. After all, you cannot lay a hammer down next to a nail, return in one week and expect the nail to be hammered into a piece of wood. 🙂
One of the smallest, are singularity cms or strapdown.js. Just mardown sites. A blog are may too complicated for most, but possible.
Nibbleblog, Yellow ([url=http://Datenstrom.se]Datenstrom.se[/url]) and PulseCMS are good points. But a real good minimal flat file blog with minimal templates are lost.
ospari arent something for me because its not free anymore and the pay what u want are a lie. the dev didnt answer and so on. after all, a blog with staceyapp would be cool. No Node.js or other crap. Sure, its looks amazing but arent possible with normal server hosting.
Any idea?
These CMS ARE for developers. The reasons why we need them are to more easily add custom functionality on top, remove the “bloat” from loading code you don’t need every page load (thereby slowing it down) and prevent plugin conflicts that comes from too many modules that offer drag-and-drop functionality during development phase.
Many of these are made for web agencies or freelancers who need to provide a more powerful set of custom features for clients than what platforms like WordPress can provide. I honestly don’t see why flat file sites will appeal to end users because they don’t read HTML anyway. How the files are structured on the back-end should not matter to end consumers.
Readers of this article might be interested in the Web Ahead episode #54 ([url=http://5by5.tv/webahead/54)]http://5by5.tv/webahead/54)[/url] on CMS-less websites. The guest is from Development Seed, who designed [url=http://HealthCare.gov]HealthCare.gov[/url] to run on the Jekyll platform. While [url=http://HealthCare.gov]HealthCare.gov[/url] is not exactly a shining example of a successful website launch (ahem), it appears that many of the problems with the site were not caused by the Jekyll components themselves ([url=https://www.ostraining.com/blog/webdesign/healthcare-gov/)]https://www.ostraining.com/…[/url]. I’m intrigued by the prospect of taking a huge number of rank-and-file employees and teaching them how to update flat files in Markdown for a databaseless CMS.
I agree that these new flat file CMSes are not really speaking to end users’ interests and desires. I build WordPress websites for small businesses, artists, and nonprofits, and the main attraction of WordPress to them is the WYSIWYG GUI where they can enter content (using their MS Word skills) and quickly publish web pages. They’re not interested in learning Markdown so they can shave a few milliseconds off their page load times. They like the simplicity of entering content in WordPress (including niceties like the mobile app) and how easy it is to add powerful features and themes via simple plugins. This current crop of CMS-less systems don’t really meet those needs.
One of my favorite blogs, [url=http://hackeducation.com]hackeducation.com[/url], recently migrated from WordPress to Jekyll, and consequently lost the “search” functionality for the site. That seriously impedes the user experience for readers, and it’s just one of the bundled features of WordPress that we take for granted as a part of a “full featured” modern webapp.
What a great comment, thanks Ted! Yes, search is a real beast to set up with many flat file sites.
Flat File sites are so focused on only managing files that they sacrifice almost everything a CMS should provide ie. the things that WordPress does. If anything, I would say that the term “flat-file” only appeals to ordinary users, and is a turn off to Developers. Hence the extra “outreach” to developers. It is interesting to me that Ted points out his clients prefer using WYSIWYG over Markdown, when the whole purpose of Markdown is to address the issues of WYSIWYG for the non-coders among us. Hence the surge of flat-file, markdown-only CMS’s.
This is the very disparity that I have been working on for years to solve. As a developer I hate WordPress, but I want my clients to be able to easily manage their websites, which is the siren song of WordPress. My solution was to create [url=http://bootpress.org/]BootPress – A Flat File CMS[/url] that utilizes SQLite to put all of the pieces together, so to speak, so that “ordinary users” can have their cake and eat it too.
I think the best thing about Markdown is that it forces the user to stop worrying about the way their content LOOKS and just focus on what it says — trusting your CSS to style it appropriately. This solves lots of problems for readers and for admins, but writers/designers are accustomed to more visual control that Markdown gives them.
They don’t understand why you need to separate presentation and semantic code on the web — they just know they want that effing word to be pink and in 15 point font, dammit! They think visually about their content (like they’ve been trained to do with MS Office and other publishing tools), and they don’t like learning some weird non-standard syntax, no matter how easy it actually is.
And honestly, there is a middle ground between text and page layout that Markdown just doesn’t have great answers for yet. A web client who wants to add UI elements like columns, boxes, tabs, etc. to their page layouts is SOL with Markdown — unless they futz with something like [MarkdownUI]([url=https://twitter.com/tedcurran/status/621320370990133248)]https://twitter.com/tedcurr…[/url]. Again, that basic user who just wants to put in a photo slider is better off using a WordPress plugin like Shortcodes Ultimate than trying to figure out a workaround in a Markdown CMS.
It is completely inconceivable to me that you have left out Harp ([url=http://harpjs.com]harpjs.com[/url]). Most people refer to ‘static’ over ‘flat file’ when having this discussion. Besides, being popular and being the best solution are not the same. For example, watch over the next few years at Nodejs completely destroys PHP and Ruby options. Adoption is important, but what is most important is adoption by people that matter. The rest of the sheep eventually follow. This was what drove the PHP adoption in the 90s. (I was there.)
Steve are you absolutely right on this one!
All those developers should get out of there basements and ask their Mom’s to use any of these so called “easy to use” CMS solutions. I have many, many clients that are your average Joe, the Tennis Teacher “I need a simple website that I can maintain myself”. It is amazing that there is no simple CMS system capable of serving those needs. The first Mambo version came closest I think. It should be whatever-file system, easy to (auto) update, normal non-“slug” language, and in-page editing.
Waiting for the first one to reply WordPress has all what I ask for… Haha… Have you ever asked your mom to maintain a WordPress website. Try explaining the difference between posts and pages…. We need Ipad-like simplicity!
Well put, Peter
I think the market is already voting here. People want ease-of-use but they aren’t looking to flat-file systems, they’re looking to SaaS solutions like Wix and Squarespace.
There’s nothing wrong with flat-file systems and they’re great for many uses … they’re just never going to be widely popular.
Thank you so much Steve for those SaaS solutions that honestly I have not yet seen before. Just checked out a demo of Squarespace and indeed it is very much in the right direction and very impressive. Let us hope that these solutions continue to build on stability instead of stacking functions on functions asked for by so called users.
My “Joe The Tennis Teacher” could very well be the target audience for such software. I could install it for him and do the initial setup and he could very soon maintain every aspect of his website… This is becoming an interesting year after all…
What about Void : [url=http://www.thisisvoid.org]www.thisisvoid.org[/url] ? Probably the lightest NoDB / flat file one
Funny that you think these platforms won’t become popular because they appeal to web developers and designers. Surely those are the people who actually build the most web sites and advise others on what systems to use. And non web debs do care that they don’t need to use a database.
Ha. I would really like to see any non developer install Ghost. It’s awesome but seriously. You think a non dev will use install ghost on a server instead of something like pico which only requires some folders and files?
Sitemagic CMS is certainly worth mentioning. It is probably the most capable Flat-File CMS on the market with a really good Designer and Editor built-in. The system is fully extensible and creating themes (templates) is really easy. [url=http://sitemagic.org]http://sitemagic.org[/url]
Great to see love in this comments thread for Pulse CMS – thank you 🙂
I’ve been looking at PineGrow. It doesn’t bill itself as a flat-file system, but if you have a set of html files, you can use it to make sitewide changes (master pages, reusable content blocks, sitewide search and replace) which make managing a website easier. We are still experimenting with it on a 1000 page website to see if it can measure up.